IRCloggy #git 2014-01-07

Logs Search ←Prev date Next date→ Channels Documentation

Provider of IRC logs since 2005.
WARNING: As Freenode became unjoinable and lost all warnings in topics, we cannot log channels on Freenode anymore.

2014-01-07

hays_ left00:00
hays_ joined00:00
girishr_ joined00:00
nafg__ joined00:00
azyr joined00:00
de-Vri-es_ joined00:00
tang^^ left00:00
skorgon_ left00:01
Petazz_ joined00:02
yhager_ joined00:02
daivyk_ joined00:02
brandonw1 joined00:02
petey left00:02
daivyk left00:02
_Qman left00:02
Octalot left00:02
raniere left00:02
mizu_no_oto left00:02
yhager left00:02
BiggFREE left00:02
netcarver left00:02
pheaver left00:02
simesy left00:02
Gamecubic left00:02
aartist left00:02
Chaser left00:02
ivan` left00:02
Will| left00:02
girishr left00:02
rafaqueque left00:02
akiva left00:02
michaelpjohnson left00:02
bwreilly left00:02
hyperair left00:02
mmattice left00:02
rpg left00:02
acharlieh left00:02
tang^ left00:02
keller left00:02
de-Vri-es left00:02
blahdodo left00:02
Layke left00:02
jxport left00:02
Lunatrius left00:02
dsantiago left00:02
gdrc left00:02
lifelike left00:02
obert left00:02
nafg left00:02
qsusOFFcs left00:02
regreddit left00:02
azyr_ left00:02
hays left00:02
Petazz left00:02
odigity left00:02
brandonw left00:02
ribasushi left00:02
Lunatrius`Lunatrius00:02
Eryn_1983_FL left00:03
mizu_no_oto joined00:03
akiva joined00:03
bwreilly_ left00:03
acharlieh joined00:03
netcarver joined00:03
gdrc joined00:03
simesy joined00:03
qsusOFFcs joined00:03
lite_ joined00:03
Will| joined00:03
Layke joined00:04
Layke left00:04
Layke joined00:04
decotii left00:04
odigity joined00:04
ribasushi joined00:04
mmattice joined00:04
ivan` joined00:04
michaelpjohnson joined00:04
pheaver joined00:04
raniere joined00:04
regreddit joined00:04
gelardi_ joined00:04
amortimer_mbp left00:04
dsantiago joined00:05
raniere left00:05
skorgon_1 joined00:05
zhost left00:06
Chaser joined00:06
blahdodo joined00:06
daube joined00:06
jxport joined00:06
halu left00:07
BillyJack joined00:07
lite_ left00:08
aton` left00:08
senj left00:09
shawndumas left00:09
leeN left00:09
xpirator joined00:10
ariscop left00:11
skorgon_1 left00:11
slyfox left00:11
dsdeiz joined00:12
rlubke left00:13
gelardi_ left00:13
MacWinner left00:13
MrSmurf joined00:13
kevwil left00:13
xSmurf left00:14
MrSmurfxSmurf00:14
royho joined00:14
rfoust joined00:14
gdogrady left00:15
mribeirodantas joined00:15
skorgon_1 joined00:15
royho left00:15
angrytongan left00:16
Landshark753 left00:17
gdogrady joined00:19
Noldorin left00:19
LuxuryMode joined00:19
joelm_ left00:19
rlubke joined00:20
abbycabs left00:20
Noldorin joined00:20
joelm joined00:20
skorgon_1 left00:20
dfanjul joined00:21
LuxuryMode left00:22
LuxuryMode joined00:22
JGarrido left00:22
ciampix left00:22
abbycabs joined00:23
abbycabs left00:23
JustinMac joined00:23
LucasKA left00:24
rafaqueq_ left00:24
skorgon_1 joined00:24
rlubke left00:24
bwreilly joined00:24
dyng joined00:26
joelm left00:26
offby1 *shudder*00:27
joelm joined00:27
offby1 nickoe: it might depend on exactly how you invoked it. Can you paste the command line?00:27
nickoe offby1: I did like "git svn clone http://my-project.googlecode.com/svn -s my_project"00:28
or rather "git svn clone file:///afs/foo/svn"00:29
offby1 what does "svn ls file:///afs/foo/svn" say?00:29
I'm wondering if you need to specify another component in that URL00:29
Note that I haven't run git-svn in _years_00:30
skorgon_1 left00:30
nickoe hmm, wait a minute00:31
seems to work now00:31
zz_karupanerurakarupanerura00:31
nickoe doing git svn clone file:///afs/foo/svnrepo00:31
tmcmahon joined00:32
smashwilson left00:32
offby1 guess you musta mistyped "svn" for "svnrepo".00:32
Or else you're pretty sloppy with what you're pasting into the channe00:33
l00:33
nickoe hmm, so when I invoke it with any options it fails, and only leave an empty00:33
offby1: no the repodir is correct, and the same as when I invoke it with or withour paramaters.00:34
skorgon_1 joined00:34
Kayra joined00:34
schemanic left00:34
zack6849 joined00:35
offby1 glad it's working00:35
be prepared to wait a _long_ time :)00:35
john2x joined00:35
marhaban left00:35
nickoe offby1: the repo is not that big only about 90 commits00:36
kukks joined00:37
nickoe Good that worked, though the log is messy because of those svn-id's and the author is not good00:37
Sonderblade left00:37
nickoe ok, seems like I can use --no-metadata and --authors-file but apparently not -s00:38
that was the option I would have expected to absolutely not make any problems00:38
offby1 nickoe: if you're not satisfied with the resulting git-repository, take a look at Eric Raymond's "reposurgeon"00:39
skorgon_1 left00:39
nickoe offby1: thank you I will keep that in mind. But now lets see if it is good enough with the authors file and no metadata.00:40
tmcmahon left00:41
nickoe ohh, I misunderstood -s00:41
:o00:41
stupid me00:41
this repo does not have trunk et.al00:41
Logs seems good now, so it appears it was just me that was the entire problem. sigh00:43
skorgon_1 joined00:43
bwreilly left00:45
phipes left00:46
cebor left00:47
hwrdprkns left00:47
johnkevinmbasco joined00:47
trevlix left00:47
Diemuzi left00:49
skorgon_1 left00:49
Cephalostrum joined00:49
bwreilly joined00:49
lavaman left00:51
LucasKA joined00:52
BiggFREE joined00:52
hwrdprkns joined00:52
phipes joined00:53
skorgon_1 joined00:53
angrytongan joined00:53
konadr joined00:55
Kayra left00:55
nickoe offby1: huh, the git repo is twice the size of the svn00:56
whilst the checked out svn is a little over twice the git00:57
refried_ left00:58
refried_ joined00:58
ariscop joined00:58
skorgon_1 left00:59
HelloShitty left00:59
ni__ joined00:59
harish_ left00:59
harish joined01:00
konadr left01:00
tvw left01:00
leeN joined01:00
leeN left01:00
offby1 nickoe: I think "git gc" or "git repack" will compress it nicely01:01
not sure which is best; rtfm01:02
oriba left01:02
mallet joined01:02
offby1 typically a git repository that's been converted from svn will be much _smaller_ than the original, not bigger.01:02
skorgon_1 joined01:03
robbyoconnor left01:03
gsvolt left01:03
phipes left01:03
robbyoconnor joined01:03
nickoe Well the "checked out" git repo is smaller than the svn checkout, but the repo on the server is smaller. I will look at gc and repack01:03
imachuchu left01:04
nwertman left01:04
pikaren joined01:04
adambrenecki joined01:04
LucasKA left01:05
d0liver test01:05
daube left01:05
dsdeiz left01:05
nickoe d0liver: pong01:05
sakb joined01:06
nevyn left01:07
hyperair joined01:07
skorgon_1 left01:08
xpirator left01:09
harish left01:11
Skrypter left01:11
Zarthus left01:12
ni__ left01:12
HelloShitty joined01:12
skorgon_1 joined01:12
arizuka joined01:13
becks__ left01:13
Skrypter joined01:13
Skrypter left01:13
sgu joined01:13
Skrypter joined01:14
sgu Hi all, I am using Magit. Is there a way to do git check -- <file> to discard changes?01:14
offby1 you meant "checkout", and yeah, I think so01:14
sgu yes.01:15
offby1 sgu: better to ask in #emacs, if you can stand the low s/n :-)01:15
rjc joined01:15
offby1 I think k, in the *magit-status* buffer, will "kill" your changes01:15
effectively doing "git checkout --"01:15
aleng_ left01:16
nevyn joined01:16
LucasKA joined01:17
Guest52001 left01:17
koda joined01:18
skorgon_1 left01:18
Skrypter left01:19
sgu yes, "k" does it.01:20
Skrypter joined01:21
Skrypter left01:21
sgu But can I select multiple files in magit and then press only one "k", and undo all the changes?01:21
skorgon_1 joined01:21
Skrypter joined01:22
Skrypter left01:22
Skrypter joined01:22
Skrypter left01:22
Skrypter joined01:23
Gamecubic_ left01:23
thalisson_ left01:23
FatDarrel left01:23
thalisson_ joined01:24
refried_ left01:24
Spudster_away joined01:24
phipes joined01:25
ehmatthes joined01:25
Spudster_away left01:25
rjc left01:25
nwertman joined01:25
dreiss left01:26
Spudster joined01:27
skorgon_1 left01:27
dragon3_ joined01:27
tobias__ left01:29
bwreilly left01:30
ehmatthes left01:30
rhubarb left01:30
skorgon_1 joined01:31
weiyang joined01:32
hahuang65 left01:34
sgambino joined01:34
Targen joined01:34
phipes left01:34
Noldorin left01:35
skorgon_1 left01:36
blinkdesign left01:37
SocalNick left01:38
LulzonAwayLulzon01:39
_Qman joined01:39
dinoangelov_ joined01:40
Targen_ joined01:40
Gamecubic_ joined01:40
skorgon_1 joined01:40
_Qman left01:41
weiyang left01:41
Targen left01:41
cads joined01:42
_Qman joined01:42
harish joined01:42
ehmatthes joined01:43
sakb left01:43
dinoangelov_ left01:44
bwreilly joined01:44
harish left01:45
redrocket joined01:45
skorgon_1 left01:46
koda left01:46
moondoggy joined01:46
acharlieh left01:46
Cyb3r-Assassin left01:46
cmpera joined01:47
acharlieh joined01:47
harish joined01:47
AaronMT left01:49
JGarrido joined01:50
skorgon_1 joined01:50
bodom joined01:50
Spudster left01:50
johnkevinmbasco left01:51
dragon3_ left01:51
dragon3_ joined01:52
luc_ joined01:52
damccull_damccull01:52
Cyb3r-Assassin joined01:52
johnkevinmbasco joined01:53
becks__ joined01:53
floatingpoint joined01:53
hwrdprkns left01:55
yhager_ left01:55
Kayra joined01:55
skorgon_1 left01:55
albel727 left01:56
dragon3_ left01:56
weiyang joined01:58
harish left01:58
thalisson_ left01:59
pbrobst_ joined01:59
ketas left01:59
skorgon_1 joined01:59
pbrobst left02:00
pbrobst_pbrobst02:00
Kayra left02:00
thalisson_ joined02:02
phipes joined02:02
mizu_no_oto left02:03
thalisson_ left02:04
pbrobst left02:04
jhulten left02:04
skorgon_1 left02:04
thalisson_ joined02:04
thalisson_ left02:04
JustinMac left02:06
pen left02:06
refried_ joined02:07
pen joined02:07
skorgon_1 joined02:08
JGarrido left02:09
Noldorin joined02:10
ryez joined02:11
pen left02:11
GT4066 joined02:13
phipes left02:13
skorgon_1 left02:14
dragon3_ joined02:14
Kayra joined02:15
daito joined02:16
JSharpe2 joined02:16
sgu left02:16
ncthom91 left02:16
skorgon_1 joined02:18
echo-area joined02:19
yacks joined02:20
Kayra left02:20
akiva left02:22
d33tah left02:22
endiruna left02:22
xSmurf left02:23
cads left02:23
dreiss joined02:23
BiggFREE left02:23
skorgon_1 left02:23
BSaboia left02:24
robbyoconnor left02:24
rasto left02:24
arizuka left02:25
Kabaka left02:25
hwrdprkns joined02:26
daivyk_ left02:26
skorgon_1 joined02:27
rurufufuss joined02:27
refried_ left02:27
robbyoconnor joined02:28
Mars___ left02:28
johnkevinmbasco left02:29
harish joined02:29
hyperair left02:29
bl33 left02:31
Gamecubic_ left02:32
Gamecubic_ joined02:32
becks__ left02:32
arizuka joined02:32
skorgon_1 left02:33
johnkevinmbasco joined02:33
Targen_ left02:34
Goplat joined02:35
veebers left02:35
veebers joined02:35
chrisss404 left02:35
SHyx0rmZ joined02:36
yacks left02:36
skorgon_1 joined02:36
dxdt left02:37
rasto joined02:37
Mars__ joined02:37
Mars__ left02:37
Gamecubic_ left02:38
acharlieh left02:39
dreiss left02:39
JSharpe2 left02:39
JustinMac joined02:40
acharlieh joined02:40
pheaver left02:40
chrisss404 joined02:41
skorgon_1 left02:42
yacks joined02:42
endiruna joined02:42
rpg_ left02:43
bitkiller_ left02:43
fire joined02:44
johnkevinmbasco left02:44
skorgon_1 joined02:45
MistuhKurtz left02:47
cilly left02:47
DrShoggoth left02:47
hyperair joined02:48
johnkevinmbasco joined02:50
Gamecubic_ joined02:50
dfanjul left02:51
skorgon_1 left02:51
cilly joined02:51
acharlieh left02:52
master_j joined02:54
skorgon_1 joined02:54
LuxuryMode left02:56
konadr joined02:56
cads joined02:56
looper left02:57
robbyoconnor left02:58
robbyoconnor joined02:58
JustinMac left02:59
skorgon_1 left03:00
konadr left03:01
skorgon_1 joined03:04
maroloccio2 left03:05
dxdt joined03:05
cads left03:05
johnkevi_ joined03:06
johnkevinmbasco left03:06
croaky left03:07
graphitemaster left03:07
graphitemaster joined03:07
chamunks joined03:09
skorgon_1 left03:09
Skrypter left03:09
croaky joined03:10
Targen joined03:10
Hoksilato joined03:11
JorDunn left03:13
skorgon_1 joined03:13
hwrdprkns left03:13
emmajane[emmajane]03:14
Kayra joined03:15
dyng left03:16
dyng joined03:16
skorgon_1 left03:18
dxdt left03:19
Kayra left03:20
m0viefreak left03:20
DolpheenDream_ joined03:20
dyng left03:21
sebastianmarkow left03:22
Dougie187 left03:22
skorgon_1 joined03:23
sebastianmarkow joined03:24
nw8l left03:24
fire left03:25
jhulten joined03:25
pen joined03:26
joelm left03:26
jerius_ joined03:28
skorgon_1 left03:28
fire joined03:28
weiyang left03:29
mdiers left03:29
ojacobson left03:29
mdiers joined03:30
bwreilly left03:31
skorgon_1 joined03:32
Mars__ joined03:32
freestyl3r left03:32
sgambino left03:34
bl33 joined03:34
rryoumaa joined03:34
Kayra joined03:34
rryoumaa i have a local branch, which rebases on master. i'd like it to rebase on maint. is there a command to get rid of all the stuff from maint to master without getting rid of my changes?03:35
mitz joined03:36
h0bbit_ joined03:36
skorgon_1 left03:37
edwardly joined03:37
SamB rryoumaa: I think you just need your basic "git rebase --onto=maint master" ?03:37
and yes, you should totally be using magit maint03:38
jerius_ left03:38
rryoumaa SamB: i am. i am considering org maint now. but you're saying my local changes which are currently grafted on master can be grafted onto maint instead via your magic command above?03:39
SamB should, yeah03:39
rryoumaa great, thanks03:39
panpainter left03:39
SamB if it starts to go wrong you can use "git rebase --abort"03:39
LulzonLulzonAway03:40
dinoangelov_ joined03:41
skorgon_1 joined03:41
rryoumaa that command looks like it would throw everything from maint to master onto master, but i will try it :)03:42
LucasKA left03:44
dinoangelov_ left03:45
john2x left03:45
angrytongan left03:45
psyl0n left03:46
john2x joined03:46
Cubit joined03:46
skorgon_1 left03:47
weiyang joined03:47
Mars__ left03:47
dyng joined03:48
LucasKA joined03:48
Cubit left03:49
nevyn rryoumaa: is maint a ancestor of master?03:49
kukks left03:49
rryoumaa nevyn: if i understand git terminology well enough, then maint and master are ancestors of master03:50
Skrypter joined03:50
Skrypter left03:50
rryoumaa i.e. maint gets merged into master03:50
nevyn ah then no.03:50
skorgon_1 joined03:50
iamjarvo joined03:50
Skrypter joined03:50
nevyn an ancestor would mean that maint is a commit in the history of master alerady03:50
Skrypter left03:50
nevyn the rebase will be fine.03:51
Skrypter joined03:51
rryoumaa nevyn: thanks03:51
noted03:51
shruggar left03:51
iamjarvo left03:52
BlueOrg left03:52
jhulten left03:52
rryoumaa after a pull, it worked. SamB nevyn thanks03:53
SamB no problem03:53
dyng left03:54
Orbitrix left03:54
rryoumaa hmm, actually it seems to have made no difference03:55
maroloccio joined03:55
skorgon_1 left03:56
mribeirodantas left03:57
rryoumaa sometimes i just want to point to my changes, select them, and say "ok, graft them onto maint now"03:57
jkremser left03:57
Cache_Money joined03:58
fire left03:58
Cache_Money I just restarted my Ubuntu VM (Win7 host) and now when I run "$ git branch" I get this error: "fatal: Failed to resolve HEAD as a valid ref."03:59
ehmatthes left04:00
Cache_Money I just checked out a new branch (titled, 'dual-shot-coords') before restarting the VM. How would I go about fixing this?04:00
milki try checking out a branch again04:02
rryoumaa is there a way to do "rebase A..B onto C"?04:03
Cache_Money milki: I can successfully checkout master but get an error when checking out 'dual-shot-coords'04:03
milki rryoumaa: sounds like cherry-pick04:04
Cache_Money: maybe that branch is now invalid then04:04
Cache_Money error^: "warning: ignoring broken ref refs/heads/dual-shot-coords." "error: pathspec 'dual-shot-coords' did not match any file(s) known to gi."04:04
milki ya...04:04
dougsland left04:04
Cache_Money so just delete it and recreate it?04:04
dreiss joined04:05
dougsland joined04:05
yhager joined04:05
Cache_Money milki: If I try to delete it it throws an error (branch not found) but still shows up when I run "$ git branch"... odd04:06
jhulten joined04:07
milki you can probably jsut do a branch -f04:07
Cache_Money milki: I've never used branch -f before.. Is this correct? "$ git branch -f master dual-shot-coords"?04:10
robbyoconnor left04:10
robbyoconnor joined04:11
cmpera left04:13
phipes joined04:13
LucasKA left04:13
master_j left04:14
dreiss left04:16
JosephFerano left04:16
JosephFerano joined04:17
Noldorin left04:17
arvind_khadri joined04:17
rryoumaa SamB: just want to confirm. you're saying that if i have 1 2 3=maint 4=master 5 6 7 8, and 5 6 7 8 are my changes, then your command will make it look like 1 2 5 6 7 8?04:18
netcarver left04:19
rryoumaa 8=local04:19
grr, let me do it right:04:19
SamB: just want to confirm. you're saying that if i have 1 2 3=maint 4=master 5 6 7 8=local, and 5 6 7 8 are my changes, then your command will make it look like 1 2 3=maint 5 6 7 8=local?04:19
LuxuryMode joined04:20
SamB rryoumaa: should do04:20
mdhopkins joined04:20
SamB if it doesn't work right there's always the reflog ...04:20
daito left04:20
rryoumaa and i do this when i am in local?04:20
SamB yeah04:21
mr_red left04:22
kcj joined04:22
Orbitrix joined04:23
Orbitrix left04:23
Orbitrix joined04:23
rryoumaa ok, i changed it so that i am auto-merging onto maint instead of master, and it is hitting me with a whole bunch of conflicts. i don't know why there would be any conflicts though, because my local chjanges don't touch any files that are conflicting.04:24
offby1 auto-merging? Isn't that POISON? (Seriously, it sounds risky)04:24
rryoumaa (and i don't know which is <<<<<<< and which is >>>>>>>>)04:24
SamB !repro04:25
gitinfo Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session -- or, even better for complex issues, design a minimal case in which your problem can be reproduced, and share it with us. This will help immensely with troubleshooting.04:25
rryoumaa offby1: i followed the normal instructions for auto-rebasing local onto master, so that when i pull i get my changes on top of master every time. that works well. but now i want to swtich to maint and do the same thing.04:25
bodom left04:27
dreiss joined04:28
dyng joined04:28
offby1 never heard of auto-rebasing, either04:29
mr_red joined04:30
gdogrady left04:31
rryoumaa SamB: basically it is just http://sprunge.us/GTBK . forget what i said about changing the auto merge thing. here i am going with what i normally have (i.e. local always rebases onto master as always) and this is the result.04:31
jhulten left04:31
rryoumaa offby1: it is in the org faq i think04:31
jcvfen left04:32
dyng left04:32
Kayra left04:32
SamB rryoumaa: you probably don't want to do that after having rebased local from master --onto=maint ...04:33
or, wait, you're not even doing a pull so why mention auto-rebase?04:33
rryoumaa right04:33
so forget that part, unless it is relevant04:33
(which i am too ignorant to determine)04:34
SamB I assume there's a difference in this file between master and maint04:34
rryoumaa ok, so why can't it just choose maint?04:34
i want to discard everything having to do with master04:34
SamB your "ob-tangle: Silence byte-compiler" doesn't touch that file ?04:35
rryoumaa that is not mine04:35
i made no such change04:35
SamB oh.04:35
hmm. I guess this was not the right command :-(04:36
rryoumaa i figured this should work without any conflicts04:36
preflex_ joined04:36
acharlieh joined04:37
preflex left04:37
preflex_preflex04:37
rryoumaa SamB: dunno, i have 1 2 3=maint 4=master 5 6 7 8=local (where 4 merges maint into master) and i want 1 2 3=maint 5 6 7 8=local (where master has nothing to do with anything).04:37
pen left04:38
SamB so, first "git rebase --abort" and then "git log --oneline --decorate HEAD master" and paste the interesting part?04:39
Cache_Money left04:39
gdogrady joined04:39
Mars__ joined04:39
Cache_Money joined04:39
pheaver joined04:39
Cassiel joined04:39
yhager left04:41
petey_ left04:41
petey joined04:42
solrize joined04:43
solrize how do i list all the branches in a repo? git branch --list doesn't do it04:43
irqq joined04:45
petey left04:46
solrize oh, --list -a04:46
sigh04:46
offby1 you can get all the refs with git for-each-ref, if that's interesting04:48
tplaner left04:48
ekoekoekoeko joined04:48
Xiti left04:49
rethus joined04:49
daito joined04:51
rryoumaa SamB: http://sprunge.us/jCWc04:51
dreiss left04:51
Orbitrix left04:52
Cache_Money $ git branch shows 2 branches (master, dual-shot-coords). WHen I run $ git branch -D dual-shot-coords it gives this error, "error: branch 'dual-shot-coords' not found"04:52
How do I delete this branch?04:52
solrize offby1, thanks, i didn't know about that. i'm not sure if it's useful but i guess it's interesting.04:52
offby1 eh04:53
heh04:53
that's my specialty04:53
SamB rryoumaa: ah04:53
"master" is just stale04:54
"git rebase --onto=origin/maint origin/master" would use the upstream tracking branches instead04:54
offby1 *sigh* as usual I forgot to push at work.04:54
SamB offby1: this is why you run sshd and leave the system on04:55
offby1 SamB: I do. But I gotta VPN in. And I'm too lazy to run the vpn client.04:55
SamB hehehe04:56
offby1 somehow it feels ... onerous04:56
konadr joined04:57
saurabh joined04:57
saurabhGuest4448504:57
ehmatthes joined04:57
rryoumaa SamB: will try tomorrow. thanks. then i presume before i do a pull i just change the auto rebase thing.04:58
SamB rryoumaa: yeah04:58
perlstein left04:58
Orbitrix joined04:59
Orbitrix left04:59
jrobichaud left04:59
daito left04:59
Orbitrix joined04:59
Cache_Money SamB: I have a branch that I can't delete (above). Do you know how I can get rid of it?04:59
Orbitrix left04:59
SamB Cache_Money: no, I have no idea what's going on there05:01
Cache_Money haha05:01
SamB I had a buffer I couldn't kill in Emacs earlier05:01
Cache_Money okay, glad I'm not the only one05:01
konadr left05:01
Cache_Money I'll throw my question up on SO05:01
DolpheenDream_ left05:02
ehmatthes left05:02
ekoekoekoeko left05:03
offby1 had a rash that wouldn't go away05:04
gdogrady left05:05
Cache_Money SamB: I had to go into ./git/refs/heads/ and delete the file05:05
ekoekoekoeko joined05:07
ewnd9 joined05:08
fire joined05:08
Vakar joined05:09
LucasKA joined05:09
irqq left05:09
Orbitrix joined05:09
Orbitrix left05:09
Orbitrix joined05:10
Orbitrix left05:10
yacks left05:10
HicksD joined05:10
Orbitrix joined05:10
HicksD_ left05:11
nwertman left05:11
dreiss joined05:12
croaky left05:12
cilly left05:12
daito joined05:12
maroloccio left05:13
dsdeiz joined05:13
cilly joined05:14
kcj left05:14
rchavik joined05:14
eaj joined05:14
LucasKA left05:15
eaj left05:15
kcj joined05:16
TLGreg left05:18
dreiss left05:18
Vakar left05:19
alezost joined05:20
ariscop left05:21
LuxuryMode left05:21
ekoekoekoeko left05:21
CheckDavid left05:22
john__ joined05:23
ketas joined05:23
mr_red left05:24
gdogrady joined05:25
jerichodotm joined05:25
sgu joined05:25
jerichodotm does anyone know hwo to setup a debian watch file to scan a github repo if thre are no releases?05:26
sgu I set up global ignore files by `git config --global core.excludesfile '~/.gitignore'` . When I change the content of ~/.gitignore later. Do I need to rerun this command?05:26
cybersphinx joined05:26
SamB jerichodotm: it involves flaming upstream about tagging05:26
mr_red joined05:26
jerichodotm SamB, hmmm. if i gave you a repo could you give me an example?05:27
SamB jerichodotm: you open an issue and say "you should do release tags more often"05:27
echo-are` joined05:27
mutex_ joined05:28
Mars__ left05:28
jerichodotm SamB, So, its not piossible if thre are no "releases"?05:28
HicksD_ joined05:28
zizzfizzix_ joined05:28
zizzfizzix_ left05:28
zizzfizzix_ joined05:28
maroloccio joined05:28
jefferai_gone joined05:28
the_cybersphinx left05:28
jefferai left05:28
veebers_ joined05:28
de-Vri-es joined05:28
echo-area left05:28
veebers left05:28
preflex_ joined05:28
LucasKA joined05:29
SamB jerichodotm: it'd have no way of knowing that something more recent had been released05:29
or, well, committed05:29
mutex left05:29
zizzfizzix left05:29
preflex left05:29
HicksD left05:29
de-Vri-es_ left05:29
rethus left05:29
qsusOFFcs left05:29
jave left05:29
Pullphinger left05:29
CalimeroTeknik left05:29
EdGruberman left05:29
preaction left05:29
preflex_preflex05:29
robbyoconnor left05:29
CalimeroTeknik joined05:29
jave joined05:29
jerichodotm SamB, because all thats avail is the master.zip, correct?05:30
rryoumaa left05:30
EdGruberman joined05:30
qsusOFFcs joined05:30
pheaver left05:30
simesy left05:30
Entomo left05:30
cehteh left05:30
joel_falcou left05:30
r0bby_ joined05:30
Vinnie_win_f joined05:30
cbreak|evil joined05:30
Dreamer3__ joined05:31
joel_falcou joined05:31
pheaver joined05:31
simesy joined05:31
SamB jerichodotm: well, even if it spoke git it'd have no idea what might be release-worthy ...05:31
Entomo joined05:31
Nugget_ joined05:31
mdhopkins left05:31
schlumpi_ joined05:31
JrCs_ joined05:32
preaction joined05:32
jerichodotm SamB, Ok. I dont know much about git at this time. I need to have the develoepr create 'releases', correct?05:32
gypsymau1o joined05:32
jerichodotm jerichodotm, lets just assume we dont care about code quality at this time and just want the latest05:32
eletuchy_ joined05:33
veebers_ left05:33
veebers joined05:33
SamB hrmm, uscan might not be for you ...05:34
kay_____1 joined05:34
SamB anyway, maybe ask in #debian-mentors on OFTC05:34
JPohlman1 joined05:34
crashev joined05:34
niax_ joined05:34
vals_ joined05:34
mackanfk1 joined05:34
przemkov1 joined05:34
Tuna-Fish2 joined05:34
ManiaArm1Yurt joined05:34
chris2_ joined05:34
BlaDe^_ joined05:34
BlaDe^ left05:34
suy_ joined05:34
qknight_ joined05:34
d0liver__ joined05:34
dsop_ joined05:34
adhawkinz joined05:34
Gaga joined05:34
joshc_ joined05:35
adhawkinzadhawkins05:35
TheBrayn_ joined05:35
Doc-Sain1ly joined05:35
adhawkins left05:35
adhawkins joined05:35
preaction left05:36
Dreamer3__ left05:36
daito left05:36
Cassiel left05:36
edwardly left05:36
hyperair left05:36
harish left05:36
nevyn left05:36
brandonw1 left05:36
saorge left05:36
paulweb515 left05:36
bbloom left05:36
beyondcreed left05:36
aberrant left05:36
mattcen left05:36
ejnahc left05:36
el3ctron_ left05:36
diget left05:36
heroux left05:36
FabTG left05:36
nitrix left05:36
haarg left05:36
kirin` left05:36
Plouj left05:36
thegorn left05:36
bambams left05:36
boz_v1 left05:36
cryptomnesia left05:36
mateu left05:36
d1b left05:36
Hamled left05:36
rainbyte left05:36
steverandy left05:36
howitdo left05:36
akitada left05:36
milki left05:36
joar left05:36
mnemoc left05:36
z^z left05:36
cite left05:36
sente left05:36
leafwiz_ joined05:36
cite joined05:36
thegorn joined05:36
heroux joined05:36
Hamled joined05:36
el3ctron joined05:36
sente joined05:36
boz_v1 joined05:36
nitrix joined05:36
d1b joined05:36
cryptomnesia joined05:36
hyperair joined05:36
steverandy joined05:36
brandonw joined05:36
milki joined05:36
preaction joined05:36
rainbyte joined05:36
z^z joined05:36
kirin` joined05:36
FabTG joined05:36
BombAw joined05:36
daito joined05:36
howitdo joined05:36
harish joined05:36
nitrixGuest7321705:36
akitada joined05:36
aberrant joined05:36
edwardly joined05:37
akitada left05:37
akitada joined05:37
howitdo left05:37
howitdo joined05:37
akitada left05:37
akitada joined05:37
wizonesolutions_ joined05:37
cehteh joined05:37
paulweb515 joined05:37
ktosiek_ joined05:37
paulweb515 left05:37
paulweb515 joined05:37
edwardly left05:37
edwardly joined05:37
eletuchy_ left05:38
phantomcircuit_ joined05:38
Cassiel joined05:38
ewet joined05:38
jon- joined05:39
mikaelj joined05:39
Dreamer3__ joined05:39
`Nothing4You joined05:39
alexwhitman_ joined05:39
Matrixiumn joined05:39
reim left05:39
RofricTheBearded left05:39
Vinnie_win left05:39
tango_ left05:39
thelamest left05:39
Tuna-Fish left05:39
kevinfagan_ left05:39
vpl_ left05:39
BombStrike left05:39
Nevik left05:39
frew left05:39
__class__ left05:39
Turicas left05:39
alexwhitman left05:39
joshc left05:39
lyang0 left05:39
Nibato left05:39
ndim left05:39
Corey left05:39
ValicekB left05:39
Nothing4You left05:39
obiwahn left05:39
ManiaArmyYurt left05:39
manitu left05:39
JStoker left05:39
EToS left05:39
monokrome left05:39
michaelmcmillan left05:39
Caius left05:39
leafwiz left05:39
CaveJohnson left05:39
wizonesolutions left05:39
interru left05:39
Fissure left05:39
Dreamer3 left05:39
tg left05:39
tchan left05:39
Stravy left05:39
jon| left05:39
ktosiek left05:39
przemkovv left05:39
mackanfkp left05:39
Baluse left05:39
ldlework left05:39
Juri_Agarin left05:39
mikaelj_ left05:39
crashev_ left05:39
GyrosGeier left05:39
cbreak left05:39
suy left05:39
schlumpi left05:39
qknight left05:39
chris2 left05:39
adhawkins-away left05:39
tm512 left05:39
mattp left05:39
Nugget left05:39
niax left05:39
TheBrayn left05:39
tsunamie left05:39
dean|away left05:39
Psi-Jack left05:39
Evanlec left05:39
Rix left05:39
gypsymauro left05:39
flexo left05:39
dsop left05:39
awakecoding left05:39
JrCs left05:39
Doc-Saintly left05:39
d0liver left05:39
koptein left05:39
JPohlmann left05:39
phantomcircuit left05:39
esto left05:39
kay______ left05:39
jerichodotm left05:39
BombAwBombStrike05:39
JrCs_JrCs05:39
esto- joined05:39
RofricTheBearded joined05:39
echo-are`echo-area05:40
Baluse joined05:40
Nevik joined05:40
jerichodotm joined05:40
jerichodotm left05:40
jerichodotm joined05:40
eletuchy left05:40
sgu Hi all: I set up global ignore files by `git config --global core.excludesfile '~/.gitignore'` . When I change the content of ~/.gitignore later. Do I need to rerun this command?05:40
mattcen joined05:40
ValicekB_ joined05:40
ewet left05:40
_class_ joined05:40
ldlework joined05:40
joar joined05:40
flexo joined05:40
vpl joined05:40
tsunamie joined05:40
Stravy joined05:40
GyrosGeier joined05:40
Guest73217 left05:40
Guest73217 joined05:40
JStoker joined05:40
JStoker left05:40
JStoker joined05:40
Tommy[D] joined05:40
awakecoding joined05:40
Guest73217nitrix05:40
phantomcircuit_phantomcircuit05:40
Nibato joined05:40
saorge_ joined05:40
koptein joined05:41
ewet joined05:41
Plouj- joined05:41
nevyn joined05:41
`Nothing4YouNothing4You05:41
gkatsev joined05:41
rurufufuss_ joined05:41
Psi-Jack joined05:41
dinoangelov_ joined05:42
gkatsev left05:42
interru joined05:42
ejnahc joined05:42
bambams joined05:42
rurufufuss_ left05:42
haarg joined05:42
tg joined05:42
kevinfagan joined05:42
rainbyte left05:43
dean joined05:43
dsdeiz left05:43
Tommy[D]_ left05:44
Than joined05:44
_class___class__05:44
mnemoc joined05:44
Rix joined05:45
tchan joined05:45
lyang0 joined05:45
ndim joined05:45
manitu joined05:45
thelamest joined05:45
Turicas joined05:46
Caius joined05:46
obiwahn joined05:46
bbloom joined05:46
EToS joined05:46
dinoangelov_ left05:46
Evanlec joined05:46
CaiusGuest9296605:47
EugeneKay No.05:47
macben joined05:47
gdogrady left05:47
daito_ joined05:47
diget joined05:48
hahuang65 joined05:48
monokrome joined05:48
Fissure joined05:48
dragon3__ joined05:48
Gooder joined05:48
ktosiek joined05:48
michaelmcmillan joined05:49
esto joined05:49
mattp joined05:49
acharlie_ joined05:50
whatitdo joined05:50
Nugget joined05:51
echo-are` joined05:51
ewet_ joined05:51
dean|away joined05:51
m-o-s joined05:51
hyperair left05:51
hyperair joined05:51
kevinfagan_ joined05:51
Doc-Saintly joined05:51
bambams_ joined05:52
bambams_ left05:52
bambams_ joined05:52
pikarenx joined05:52
cilly left05:52
Than left05:52
tg left05:52
bambams left05:52
kevinfagan left05:52
dean left05:52
ewet left05:52
GyrosGeier left05:52
mattcen left05:52
Baluse left05:52
esto- left05:52
phantomcircuit left05:52
ktosiek_ left05:52
daito left05:52
howitdo left05:52
ValicekB_ left05:52
Doc-Sain1ly left05:52
mackanfk1 left05:52
vals_ left05:52
gypsymau1o left05:52
Nugget_ left05:52
qsusOFFcs left05:52
jave left05:52
echo-area left05:52
maroloccio left05:52
cybersphinx left05:52
john__ left05:52
acharlieh left05:52
dougsland left05:52
Skrypter left05:52
weiyang left05:52
h0bbit_ left05:52
Gamecubic_ left05:52
endiruna left05:52
chrisss404 left05:52
rasto left05:52
dragon3_ left05:52
GT4066 left05:52
floatingpoint left05:52
_Qman left05:52
pikaren left05:52
BillyJack left05:52
Chaser left05:52
blahdodo left05:52
jxport left05:52
dean|awaydean05:52
Xiti joined05:52
the_cybersphinx joined05:52
cillybabe joined05:52
phantomcircuit joined05:52
jave joined05:53
kini joined05:53
Skrypter joined05:53
weiyang joined05:53
BillyJack joined05:54
dougsland joined05:54
chrisss404 joined05:54
maroloccio joined05:55
m-o-s left05:55
aberrant left05:56
jxport joined05:57
hwrdprkns joined05:57
tg joined05:57
qsusOFFcs joined05:57
GyrosGeier joined05:57
mattcen joined05:57
Baluse joined05:57
mackanfk1 joined05:57
gypsymau1o joined05:57
h0bbit_ joined05:57
Gamecubic_ joined05:57
endiruna joined05:57
rasto joined05:57
_Qman joined05:57
Chaser joined05:57
blahdodo joined05:57
Kabaka joined05:57
ValicekB joined05:58
tgGuest4429405:58
tango_ joined06:01
jizanthapus joined06:01
croaky joined06:02
yhager joined06:03
ryez left06:04
nw8l joined06:04
GeertV joined06:04
xSmurf joined06:06
ariscop joined06:10
bambams_bambams06:11
nicedice left06:11
nicedice joined06:12
ahmedelgabri joined06:13
ahmedelgabri left06:13
travisstaton joined06:13
ahmedelgabri joined06:13
johnkevi_ left06:14
fire left06:14
dev__ joined06:14
Vivekananda joined06:15
johnkevinmbasco joined06:17
dreiss joined06:17
tpollard left06:17
veebers left06:18
jizanthapus left06:18
ArcTanSusan joined06:19
netsroht left06:20
dreiss left06:21
netsroht joined06:23
sgu left06:23
donmichelangelo left06:23
donmichelangelo joined06:23
maroloccio left06:23
Kayra joined06:24
Mars__ joined06:28
Kayra left06:28
frew joined06:30
kbart joined06:31
pikarenx left06:32
pantsman joined06:32
pikaren joined06:32
hwrdprkns left06:32
Mars__ left06:33
yhager left06:33
ThomasLocke joined06:37
dsdeiz joined06:38
yhager joined06:38
the_cybersphinx left06:41
petey joined06:42
dyng joined06:42
bwreilly joined06:44
daito_ left06:46
pheaver left06:47
petey left06:47
daito joined06:48
Hoksilato_ joined06:49
ewet_ left06:50
Hoksilato left06:50
jackneill joined06:51
kay_____1kay_____06:51
kay_____kay______06:51
nw8l left06:55
matanya joined06:55
dyng left06:55
h_kon joined06:55
croaky left06:55
dyng joined06:56
stodan joined06:57
pikaren left06:57
konadr joined06:57
perlstein joined06:58
dyng left06:59
tobias__ joined06:59
LucasKA left06:59
pantsman left06:59
freimatz joined07:01
hoschi joined07:02
konadr left07:02
dyng joined07:04
mishok13 joined07:05
jceb joined07:06
nilg joined07:08
dyng left07:08
SHyx0rmZ left07:09
lite_ joined07:09
dyng joined07:09
d33tah joined07:15
acharlie_ left07:15
Kayra joined07:15
jargon left07:15
fire joined07:18
cagedwisdom left07:18
cagedwisdom joined07:20
Kayra left07:20
cagedwisdom left07:21
travisstaton left07:22
cagedwisdom joined07:22
bitslayer joined07:23
cagedwisdom left07:23
bitslayer I could use some help with git...07:24
ciampix joined07:24
mlechner joined07:26
revyd joined07:26
cbreak-work joined07:27
bitslayer anyone out there wanna help with a problem I'm having?07:27
cbreak-work left07:28
cmpera joined07:28
Layke left07:28
cbreak-work joined07:28
pen joined07:28
kay______ left07:28
takamichi joined07:28
eletuchy joined07:29
Mars__ joined07:29
eijk_ joined07:30
fire left07:31
milki bitslayer: !ask07:31
gitinfo bitslayer: Yes, it's okay to ask questions here.... in fact, you just asked one! ;-) Pretty much any question is fine. We're not terribly picky, but we might be asleep. Please be patient and you should get an answer soon.07:31
mateu joined07:31
simpleminds left07:31
shihan_ left07:31
bwreilly left07:31
LeeXiaolan left07:31
mishok13 left07:32
cagedwisdom joined07:32
cagedwisdom_ joined07:32
cagedwisdom_ left07:32
cagedwisdom left07:32
peper left07:32
cagedwisdom joined07:33
Mars__ left07:34
Hamled left07:34
quotemstr left07:34
eijk joined07:34
bitslayer left07:34
eijk_ left07:34
keeb left07:35
keeb joined07:35
shihan joined07:36
simpleminds joined07:36
Hamled joined07:36
holderbaum joined07:37
peper joined07:39
obert_ left07:39
swaechter joined07:40
quotemstr joined07:40
Kicer86 joined07:40
macben left07:42
dinoangelov_ joined07:42
simesy left07:43
pagalbanda joined07:43
pagalbanda left07:43
bitkid_bitkid07:44
solrize left07:44
kay______ joined07:46
dreiss joined07:46
macben joined07:47
dinoangelov_ left07:47
arand joined07:49
FerreusR joined07:49
hits joined07:49
hits left07:49
nnyk_ joined07:49
rendar joined07:50
m-o-s joined07:52
croaky joined07:52
MattMaker joined07:53
halu joined07:53
vurma joined07:54
arand left07:55
nilg left07:55
HelperW joined07:56
PeterBooker joined07:56
m-o-s left07:57
mishok13 joined07:58
nilg joined07:58
MrQuist joined07:58
Anticom joined07:58
juril joined08:01
harish left08:02
matanya left08:03
suy_suy08:03
matanya joined08:04
el3ctron left08:04
bartzy joined08:04
marthinal joined08:06
tm512 joined08:07
croaky left08:08
ewnd9 left08:08
Timmy joined08:09
ewnd9 joined08:09
mrhooray joined08:09
mrhooray left08:09
mrhooray joined08:10
papegaaij joined08:10
mrhooray left08:10
ewnd9 left08:11
mrhooray joined08:11
niax_niax08:12
ewnd9 joined08:12
Yonkie joined08:13
noname001__ joined08:13
Norm_ left08:15
Kayra joined08:15
harobed_ joined08:16
harobed_ left08:16
beyondcreed joined08:16
beyondcreed left08:16
beyondcreed joined08:16
harobed_ joined08:16
macben left08:18
macben joined08:18
Kayra left08:19
daito left08:19
codehotter joined08:20
codehotter How do I find out which remote git push pushes to08:20
dyng left08:21
jast depends on your configuration, most notably the value of push.default08:21
dyng joined08:21
alorence joined08:21
canton7 codehotter, from man git-push: When the command line does not specify where to push with the <repository> argument, branch.*.remote configuration for the current branch is consulted to determine where to push. If the configuration is missing, it defaults to origin.08:21
gitinfo codehotter: the git-push manpage is available at http://jk.gs/git-push.html08:21
codehotter jast how do I query a configuration value?08:21
TomyWork joined08:22
jast 'git config <name>'08:22
hits joined08:22
hits left08:22
jmcantrell joined08:22
canton7 jast, does push.default affect which *remote* to push to?08:22
codehotter canton7 ! ah, I missed the part "if the configuration is missing, it defaults to origin"08:22
jast canton7: not sure...08:22
mishok13 left08:22
jnavila joined08:22
jast well, it does in that if push.default is set to 'upstream' or similar, it doesn't push at all if there is no tracking config :)08:22
canton7 I thought at first it would, then realised it only specifies the ref on the remote, not the remote itself08:22
that's very true, however08:23
dyng left08:23
codehotter git config push.default prints out nothing08:23
dyng joined08:23
shruggar joined08:23
jast currently the default is 'matching'08:23
canton7 it defaults to 'matching', then (and will default to 'simple' in git 2.0)08:23
gua_ould joined08:23
Raging_Hog joined08:26
Kabaka left08:27
harish joined08:28
Nevik prefers it set to 'upstream'08:28
longbeach joined08:29
mishok13 joined08:29
aChatir joined08:30
Mars__ joined08:30
macben left08:30
aChatir left08:30
codehotter would this be a reliable way to get the remote "git push" would push to from bash, or is there a much easier way?08:31
git config branch.$BRANCH.remote || git remote -v | sed -ne "s/^origin\t\(.*\) (push)/\1/p"08:31
dreiss left08:32
Mars___ joined08:32
Mars__ left08:32
solrize joined08:32
dreiss joined08:32
arizuka_ joined08:33
arizuka left08:33
GeertV left08:34
dyng left08:34
GeertV joined08:34
dyng joined08:34
Spyro left08:34
albel727 joined08:35
osse codehotter: git config branch.$branch.remote || git config remote.origin.pushurl || git config.remote.origin.url08:35
but git config branch.$branch.remote gives only the name of a remote, not the url08:36
Spyro joined08:36
Timmy left08:36
soee joined08:37
smef joined08:38
daito joined08:38
harish left08:38
buzzybron joined08:39
smef left08:39
smef joined08:39
mdiers left08:40
lavong joined08:40
obert joined08:40
akiva joined08:40
Goplat left08:41
rath joined08:42
simesy joined08:43
beyondcreed left08:43
daito left08:43
petey joined08:43
bitslayer joined08:44
bitslayer anyone out there?08:45
gitinfo set mode: +v08:45
master_j joined08:45
bitslayer anyone out there?08:46
milki !ask08:46
gitinfo Yes, it's okay to ask questions here.... in fact, you just asked one! ;-) Pretty much any question is fine. We're not terribly picky, but we might be asleep. Please be patient and you should get an answer soon.08:46
dreiss left08:47
arizuka_ left08:47
bitslayer when I do a git branch -a I get * angularjsdev, master, remotes/origin/angularjsdev, remotes/origin/master08:47
Gamecubic_ left08:48
simesy left08:48
bitslayer but my angularjsdev doesn't have whats up on the remote08:48
i.e. the files don't match08:48
petey left08:48
konadr joined08:48
bitslayer do I have a local branch that is separate from the remote branch?08:48
psoo joined08:49
joaoh82 left08:50
DrHouse|Yuno left08:50
master_j left08:50
thierryp joined08:51
Maxdamantus left08:52
osse it's alway seperate08:52
gir status should tell you08:52
Maxdamantus joined08:53
shruggar left08:53
arizuka joined08:53
konadr left08:54
konadr joined08:54
johnkevinmbasco left08:55
nilg left08:55
LucasKA joined08:56
Devastator joined08:56
alexwhitman_ left08:56
daito joined08:57
alexwhitman joined08:58
nilg joined08:58
FiLoSsOfIe hello08:59
kesha joined09:00
LucasKA left09:01
daito left09:02
DanielHollands left09:02
mustafavelioglu joined09:02
SpaghettiCat_ joined09:04
HelperW left09:04
slyfox joined09:05
slyfox left09:05
slyfox joined09:05
bl33 left09:06
Kayra joined09:06
Sonderblade joined09:06
Sonderblade left09:07
jceb left09:07
Sonderblade joined09:07
dyng left09:07
dyng joined09:07
LimeBlast joined09:07
longbeach left09:08
harish joined09:08
pikaren joined09:08
Mars___ left09:09
asedeno left09:11
SonderbladeWork joined09:11
SonderbladeWork left09:11
akiva left09:11
SonderbladeWork joined09:12
Kabaka joined09:12
codehotter osse: thanks! So how's this?09:12
REMOTE=$(git config branch.$BRANCH.remote); REPOURL=$(git config remote.${REMOTE:=origin}.pushu09:12
rl || git config remote.$REMOTE.url);09:12
Keridos_offKeridos09:13
Silmarilion joined09:13
mrhooray left09:13
weiyang left09:14
croaky joined09:14
MattMaker left09:14
nnyk_ left09:14
MattMaker joined09:15
SonderbladeWork left09:15
nnyk_ joined09:16
ArcTanSusan left09:16
MattMaker left09:16
MattMaker joined09:17
marthinal left09:17
asedeno joined09:18
JanC left09:19
croaky left09:19
shruggar joined09:20
jceb joined09:21
MattMake_ joined09:23
macben joined09:24
MattMaker left09:24
joaoh82 joined09:24
WaReZ joined09:25
SpaghettiCat_ Hello, I can't understand this problem: git clone fails on cloning TO a NAS (NTFS), but clones fine to a local hard-drive (NTFS). Here is the error: http://pastie.org/private/leudedrrjiwiu9zfborssa Using git 1.8.5.2 on Cygwin. Any ideas are appreciated.09:26
m0ltar joined09:28
Skrypter left09:28
moritz SpaghettiCat_: usually the problem with network file systems is that they don't support proper locking09:29
Skrypter joined09:29
Skrypter left09:29
joaoh82 left09:29
MattMake_ left09:29
MattMaker joined09:30
moritz SpaghettiCat_: the way that problem is reported is less than awesome, though09:30
blinkdesign joined09:31
SpaghettiCat_ moritz: Thanks for the help. I suppose I should just have the repo on the local drive. Network drives seemed to me so transparent until now :/09:32
JanC joined09:32
moritz SpaghettiCat_: one of the many imperfect abstractions in the computing world09:33
bitslayer left09:34
DoniRS joined09:37
LeeXiaolan joined09:38
Kayra left09:38
daito joined09:39
macben left09:39
marthinal joined09:40
vince_prignano joined09:40
macben joined09:41
dinoangelov_ joined09:43
jarr0dsz joined09:44
simesy joined09:44
khmarbaise joined09:44
MrQuist left09:45
irqq joined09:45
hyperair left09:46
ChrisAnn joined09:46
mikecmpbll joined09:46
buzzybron left09:47
dinoangelov_ left09:48
matteo joined09:48
Norm_ joined09:49
simesy left09:49
Kabaka left09:50
kesha left09:51
bvi joined09:53
LeeXiaolan left09:55
LucasKA joined09:55
macben left09:55
macben joined09:56
nnyk_ left09:57
subhojit777 joined09:57
dragon3__ left09:57
subhojit777 left09:57
subhojit777 joined09:58
subhojit777 left09:58
subhojit777 joined09:58
jkremser joined09:59
jkremser left09:59
jkremser joined09:59
pmoust joined09:59
jceb left10:00
LucasKA left10:00
alexO joined10:01
smef left10:02
smef joined10:02
dinoangelov_ joined10:04
bl33 joined10:04
cff joined10:04
dyng left10:05
dyng joined10:05
jooools joined10:06
arietis joined10:06
dinoangelov_ left10:08
dinoangelov_ joined10:08
Kayra joined10:08
MattMaker left10:08
Guest92966Caius10:09
Caius left10:09
Caius joined10:09
dyng_ joined10:09
dyng left10:09
dyng_ left10:11
joaoh82 joined10:11
dyng joined10:11
echo-are` left10:12
macben left10:12
macben joined10:12
jceb joined10:13
vtunka joined10:13
LeeXiaolan joined10:14
Mars__ joined10:15
dyng left10:16
SpaghettiCat_ left10:17
dinoangelov_ left10:18
SpaghettiCat_ joined10:18
dinoangelov__ joined10:18
SpaghettiCat_ left10:18
ahmedelgabri left10:18
khmarbaise left10:19
Orbitrix left10:19
mdiers joined10:20
endiruna left10:23
nipennem joined10:25
nipennem left10:25
ahmedelgabri joined10:26
osse codehotter: looks good10:26
chris2_chris210:26
bandroidx left10:26
perlstein left10:26
osse codehotter: though it's a bad habit to use uppercase variable names :)10:28
danicampa901 joined10:31
dinoangelov__ left10:31
Macaveli joined10:31
bandroidx joined10:31
Guest44294 left10:32
Guest44294 joined10:32
Guest44294tg10:32
arizuka left10:32
h_kon left10:33
juliohm joined10:33
Macaveli left10:34
arizuka joined10:34
h_kon joined10:34
h_kon left10:34
Macaveli joined10:34
r0bby_ left10:36
jgoss joined10:36
arizuka left10:36
m0ltar left10:37
macben left10:38
gain joined10:38
mitz left10:38
nilg left10:39
Skrypter joined10:40
luc_ left10:41
jargon joined10:41
zapb joined10:43
takamichi left10:43
dyng joined10:44
petey joined10:44
Skrypter left10:45
tma996 joined10:46
lavong left10:48
gain left10:48
yhager left10:48
takamichi joined10:48
rafeca joined10:48
petey left10:49
HicksD joined10:49
phipes left10:49
psyl0n joined10:51
Hoksilato_ left10:51
HicksD_ left10:51
GeertV left10:52
leandrosansilva joined10:52
cjsarette joined10:53
dropdrive left10:54
mitz joined10:55
dropdrive joined10:55
m0ltar joined10:55
Macaveli left10:55
LucasKA joined10:56
thevishy joined10:57
h_kon joined10:58
Macaveli joined10:59
harobed_ left10:59
LucasKA left11:02
jackneill left11:03
reg_reg joined11:03
marhaban joined11:03
reg_reg left11:04
Anticom left11:05
Anticom joined11:07
cmpera left11:07
SWakey joined11:07
Sly1986 joined11:08
psyl0n left11:12
eletuchy left11:12
vince_prignano left11:12
Khisanth left11:13
eletuchy joined11:13
psyl0n joined11:13
marhaban left11:14
beautifulmind joined11:14
Timmy joined11:15
Orbitrix joined11:16
eletuchy_ joined11:16
Orbitrix left11:16
eletuchy left11:16
Orbitrix joined11:16
daito left11:16
RaceCondition joined11:19
Kayra left11:25
Timmy left11:26
Khisanth joined11:27
Orbitrix left11:29
HelperW joined11:30
the_cybersphinx joined11:31
leandrosansilva_ joined11:32
mustafavelioglu left11:32
leandrosansilva left11:32
mlimaloureiro joined11:33
Kayra joined11:37
dyng left11:38
davividal joined11:39
dyng joined11:39
vairi joined11:39
mustafavelioglu joined11:40
dyng left11:41
dyng joined11:41
codehotter hi, if I want to get a single branch from upstream from a large repo, and then don't care about the history of that copy (to further save space) but then I still want to be able to update that branch from upstream and download only the deltas11:43
what should I do?11:43
MaximB joined11:43
MaximB when I do git remote set-url origin http://stash-admin@x.x.x.x:7990/scm/git_test/git_test.git ;I get fatal: No such remote 'origin'11:44
Corey joined11:44
codehotter MaximB: maybe git remote add origin?11:45
macben joined11:45
MaximB codehotter: yes, there is also this command, but for some reason also the command I just used was specified11:46
mikecmpbll left11:46
mustafavelioglu left11:46
mikecmpbll joined11:46
mlimaloureiro left11:47
mvelioglu joined11:47
MaximB codehotter: so this command git remote add origin http://stash-admin@x.x.x.x:7990/scm/git_test/git_test.git will add the specified server as master?11:47
guide left11:47
crater2160 left11:47
simesy joined11:47
qwertz_ joined11:48
skorgon_1 joined11:48
lavong joined11:50
marhaban joined11:50
SWakey left11:50
alexO left11:51
simesy left11:52
Mangellan2497 left11:53
codehotter MaximB: what do you mean by 'master'?11:54
Mangellan2497 joined11:54
MaximB nvm I think I managed to do it11:54
vince_prignano joined11:55
GeertV joined11:56
Guest93929 joined11:56
FerreusR left11:56
arand joined11:57
angelsl joined11:57
Mangellan2497 left11:58
LucasKA joined11:59
Mangellan2497 joined11:59
leandrosansilva_ left12:01
leandrosansilva joined12:01
Nizumzen joined12:01
arand left12:02
paulweb515 left12:02
ToBeFree left12:03
LucasKA left12:04
shruggar left12:04
eaj joined12:04
cmpera joined12:05
ggherdov left12:06
ggherdov joined12:06
ggherdov left12:06
ggherdov joined12:06
Mangellan2497 left12:06
takamichi left12:11
kill-9_ joined12:11
KeridosKeridos_off12:11
przemkov1 left12:11
FerreusR joined12:12
daito joined12:12
Guest44485 left12:14
kriks joined12:14
dinoangelov joined12:14
doev joined12:14
irqq left12:14
irqq joined12:15
zhost joined12:15
daito left12:17
arietis left12:18
vairi left12:18
quackgyver joined12:18
przemkovv joined12:20
dinoangelov left12:22
takamichi joined12:23
gdogrady joined12:24
xlinkz0 joined12:25
perlstein joined12:26
badewwr left12:26
ConstantineXVI left12:27
ciampix left12:28
bitkiller_ joined12:28
danicampa901 left12:29
johnkevinmbasco joined12:29
ws2k3 left12:29
thiago joined12:30
przemkovv left12:33
cmpera_ joined12:33
cmpera left12:34
cmpera_cmpera12:34
ConstantineXVI joined12:35
przemkovv joined12:35
hyperair joined12:37
h0bbit_ left12:39
hyperair left12:43
ni__ joined12:43
zapb left12:44
Mangellan2497 joined12:45
jerichodotm left12:45
thierryp left12:45
fire joined12:45
thierryp joined12:46
macben left12:46
ewnd9 left12:47
freestyl3r joined12:48
hudku joined12:48
przemkovv left12:52
carbon1 joined12:52
renatow joined12:52
badessdf joined12:54
xadd joined12:54
przemkovv joined12:54
hits joined12:54
codehotter if I cloned a single branch, how do I clone another branch into that repository?12:54
jayrobot left12:55
hits left12:55
hudku left12:56
pmoust left12:56
fanf2 joined12:56
noecc joined12:56
hudku joined12:56
cannonball joined12:57
Kabaka joined12:57
msafi joined12:59
gdogrady left12:59
jast codehotter: git clones all branches by default but doesn't make them all visible in the normal list. you can just 'git checkout <name of branch>' to make it show up12:59
msafi Can I download a specific directory from a repository hosted on, for example, GitHub?12:59
jast 'git branch -r' to see what got copied during clone (with an 'origin/' prefix)12:59
gdogrady joined12:59
rfoust left12:59
LucasKA joined12:59
msafi Can I at least download a specific branch?13:01
arietis joined13:02
jast msafi: yes... https://help.github.com/articles/downloading-files-from-the-command-line13:02
I'm not sure whether downloading subdirs is supported... if it is, it doesn't seem to be documented13:02
Orbitrix joined13:03
jast the git way of doing it is disabled on github's servers13:03
Orbitrix left13:03
Orbitrix joined13:03
bremner what's the git way of doing it? git archive?13:03
jast yeah13:03
halu left13:04
LucasKA left13:04
blotchy left13:04
renatow left13:05
carbon1 left13:05
dyng left13:06
dyng joined13:07
codehotter jast: git branch -r only shows the branch I specified with my original single-branch clone (as expected)13:07
jast: how do I get one additional branch?13:07
ni__ left13:08
anli__ joined13:08
anli__ I want to apply the changes of a certain commit without creating a commit, how do I do that with tortoisegit?13:09
mdhopkins joined13:09
MattMaker joined13:09
MaximB left13:09
msafi1 joined13:09
codehotter Is it possible to make a shallow fetch just like you can make a shallow clone?13:10
(or does that question not make sense?)13:10
moritz it doesn't make too much sense13:10
a clone starts the life of a repo; there you can decide to not fetch the inital history13:11
smashwilson joined13:11
moritz but once you have a repo, you already have that decision made13:11
dyng left13:11
msafi left13:11
codehotter OK, I want to make a clone without the initial history, of only a single branch. And later, once that repo exists, I want to add a second branch to that repo, also without initial history.13:11
how can I do this?13:11
angelsl left13:12
quetz joined13:12
codehotter I did git clone --depth 1 --branch branch13:12
how to I get branch branch2 into that same repo?13:12
anli__ Was my previous posting visible? I am not sure nickserv was letting it through13:12
codehotter git 1.7.10 atm but willing to upgrade13:12
arietis left13:13
moritz codehotter: I don't have much experience with shallow clones, so I don't know, but my guess is 'not possible'13:13
codehotter anli__: your question does not make sense to me. What does it mean to 'apply the changes' of a commit? Do you want your working copy to have the changes?13:13
moritz anli__: yes, it was visible13:13
anli__ codehotter: yes, moritz: thx13:13
moritz anli__: you can cherry-pick it, and then git reset HEAD^13:13
anli__ And then I want the working copy to get changes of another commit after the first one13:13
So I must have a commit in between then?13:14
codehotter moritz: what would I do if I cloned a single branch if it wasn't a shallow clone? How would I add the other branch?13:14
moritz anli__: the git reset kills the commit13:14
anli__ But keeps the changes then?13:14
moritz yes13:14
anli__ cool13:14
shruggar joined13:15
Kayra left13:15
anli__ And I will get a dumb end then that will magically dissapear when cleaning?13:15
Kayra joined13:15
moritz anli__: and it sounds like you want to re-invent squashing of commits13:16
moritz doesn't know what a "dumb end" is13:16
anli__ No, I only want to reinvent "apply this and that commit"13:16
moritz that's git cherry-pick13:16
anli__ One that is not connected13:16
...without getting new commits13:16
Kayra left13:16
niatish joined13:16
moritz but why? in the end you'll want to have commits, no?13:17
mlimaloureiro joined13:17
moritz I mean, that's the whole point of using a version control system :-)13:17
anli__ But maybe I want to make a big commit out of them13:17
HelperW_ joined13:17
moritz that's what squashing is13:17
anli__ Oh, ok13:17
moritz combining several commits into one13:17
which you seem to be wanting to re-invent :-)13:17
ewet joined13:17
HelperW left13:17
anli__ I dont want to modify commits actually13:18
ewet left13:18
anli__ And making one commit out of two is modification, I assume13:18
moritz yes, it is13:18
pikaren how often do i need to commit13:18
Kayra joined13:18
Kayra left13:18
moritz so is cherry-picking (it modifies the parent, thus the sha1)13:18
well, creates a modified copy13:18
anli__ pikaren: every time you have done something working that you can put a name on :)13:19
ewet joined13:19
moritz anli__: https://help.github.com/articles/interactive-rebase might be of interest13:19
anli__ ok13:19
ewet left13:19
ewet joined13:20
jarr0dsz left13:20
ewet left13:20
Kayra joined13:21
anli__ When I edit a conflict in tortoisegit, the scrolling will reset to topmost, why?13:21
ewet joined13:21
moritz dunno, ask the tortoise devs :-)13:22
ewet left13:22
ewet joined13:22
anli__ I thought maybe one or two of them are in here13:22
h0bbit_ joined13:22
regreddit left13:23
mvelioglu left13:23
psoo left13:23
psoo joined13:23
mvelioglu joined13:23
mlimaloureiro left13:24
r0bby_ joined13:25
joaoh82 left13:25
HelperW_ left13:26
notjohn joined13:26
HelperW_ joined13:26
harobed joined13:27
ewnd9 joined13:29
rfoust joined13:29
dinoangelov joined13:29
r0bby_robbyoconnor13:29
notjohn left13:29
anli__ If I cherry pick and get conflicts, will a click on abort take me to the state before the cherry pick?13:30
A click of "abort" in the dialog where I resolve conflicts, that is13:30
mdhopkins left13:31
notjohn joined13:31
ahmedelgabri left13:31
decotii joined13:31
wakeup joined13:32
wakeup Hi13:32
dr_lepper anli__: probably yes13:32
anli__: !gui13:32
gitinfo anli__: Graphical user interfaces are not supported here. If you want to get support, it needs to be through the git CLI. Reasons: 1) Because very few people here use the graphical interface. 2) Because giving instructions for GUI's is difficult. 3) The command line gives you a history of what commands you have executed.13:32
wakeup I want git to set correct mtimes for checked out files. E.g. the13:32
dates when the files actually were modified.13:32
Any chance I can get what I want?13:33
ewet is this http://pastie.org/8609896 a good example for the rebase/merge controversy?13:33
macben joined13:33
sgambino joined13:33
msafi joined13:33
msafi How can I move a directory from one branch to another?13:34
dr_lepper wakeup: not with git itself, but there are third-party scripts that do something like that13:34
Entomo left13:34
wakeup dr_lepper: which, and what does somethinbg like that mean?13:34
Imho this should be the default behaviour.13:34
hudku left13:35
dr_lepper wakeup: just google around13:35
razzle joined13:35
dr_lepper git does not do it13:35
msafi1 left13:35
dr_lepper ewet: it looks like it's extremely workflow-dependent13:36
hyperair joined13:37
dr_lepper most workflows do not allow commits which break tests13:37
ewet dr_lepper: absolutely13:37
wakeup bullshit workflow then lol13:37
ewet well the commit would actuallly have happened already13:37
and test would have been fine13:38
only after the rebase things would break, wouldn't they?13:38
dr_lepper the commit message "test OK" is meaningless in such workflows, because all commits are "test OK"-commits13:38
macben left13:39
ewet dr_lepper: yeah, but that's just to explain things ...13:39
it doesn't make any difference13:40
diegoviola joined13:40
marhaban left13:40
[emmajane]emmajane13:40
looper joined13:40
breakingthings joined13:41
ToBeFree joined13:43
ewet I just needed a commit message that would become wrong13:44
Roch_ joined13:44
dr_lepper then don't describe the state, describe the changes13:46
ewet left13:46
thrakcattack left13:46
macben joined13:46
mikecmpbll left13:46
macben left13:47
ewet joined13:47
thrakcattack joined13:48
mvelioglu left13:48
ewet left13:48
tmcmahon joined13:48
amcsi_work left13:48
amcsi_work joined13:49
ewet joined13:49
psoo_ joined13:49
ewet left13:50
vince_prignano left13:50
macben joined13:51
IrishPrime joined13:51
ewet joined13:51
ewet left13:51
amortimer_mbp joined13:52
psoo left13:52
thiago left13:52
ewet joined13:52
shinnya joined13:52
tmcmahon left13:52
ewet left13:52
h0bbit_ left13:53
ahmedelgabri joined13:53
ewet joined13:54
jerius joined13:54
subhojit777 left13:55
joaoh82 joined13:56
ItSANgo left13:57
robbyoconnor left13:57
ws2k3 joined13:58
ws2k3 left13:58
ws2k3 joined13:58
Mars__ left13:58
gdogrady left13:59
Mars__ joined13:59
ni__ joined13:59
floatingpoint joined14:00
LucasKA joined14:00
ewet cleaned it up a little http://pastie.org/860996314:00
joaoh82 left14:01
jsilver left14:01
thiago joined14:02
hellopat joined14:02
razzle left14:03
paulweb515 joined14:04
DaisyTheCow joined14:04
Mars__ left14:04
paulweb515 left14:04
paulweb515 joined14:04
LucasKA left14:04
Destos joined14:05
mvelioglu joined14:06
ahmedelgabri left14:09
FortySix2 joined14:09
FortySix2 left14:10
b3tamax joined14:11
FortySix2 joined14:12
Destos left14:13
vtunka left14:14
Norm__ joined14:14
Norm__ left14:15
gfontenot joined14:16
ItSANgo joined14:16
tma996 left14:18
bandroidx left14:18
Nibato left14:18
tma996 joined14:19
hs366 joined14:19
geekbri joined14:20
bandroidx joined14:21
cff Why does Github recommend core.autocrlf true on Windows? https://help.github.com/articles/dealing-with-line-endings#global-setting14:21
and input on OS X and Linux ?14:21
nilg joined14:21
jerius left14:22
phile joined14:22
cff that means that on OS X and Linux the line ending won't be changed by Git?14:22
Barringer joined14:22
cff i.e. the files are checked out as is14:22
tobias__ left14:23
juliohm left14:23
tma996 left14:23
mdiers left14:25
jast personally I think the best way to deal with line ending conversion is the one outlined in !crlf14:26
gitinfo to fix problems with line endings on different platforms, check out http://line-endings.gitrecipes.de/14:26
Silmarilion left14:26
angelsl joined14:26
thiago left14:26
Dougie187 joined14:27
jerichodotm joined14:28
jerichodotm left14:28
jerichodotm joined14:28
xadd left14:28
tma996 joined14:29
Macaveli left14:29
crater2150 joined14:29
cff so what value should core.eol have?14:29
or it's set automatically ?14:29
jast each build of git should have a proper default14:30
the only exception is cygwin (which is on windows but pretends to be unixy, so it uses LF)14:30
Raging_Hog left14:31
kbart left14:31
vince_prignano joined14:31
vince_prignano left14:32
zumba_addict so folks, we made a separate branch that we are treating as master. We called it bug-fixes. We don't want to directly push to this branch. We will create our own local-fixes branch, have the other devs review the code, then merge to bug-fixes when 5 devs approve code. If let's say I will be making a new local-fix branch, how do I make sure that I only pull from the most recent bug-fixes remote branch?14:32
rchavik left14:33
bgerber left14:33
guide joined14:33
Macaveli joined14:33
GeertV left14:35
GeertV joined14:35
beautifulmind left14:37
soee left14:37
bgerber joined14:37
daivyk joined14:38
joshc_joshc14:38
gancl joined14:39
GeertV left14:40
Destos joined14:40
tma996 left14:40
Kayra left14:40
Barringer left14:41
pixelgremlins joined14:41
ahmedelgabri joined14:41
arietis joined14:41
macben left14:43
marhaban joined14:43
Advocation joined14:44
murosai joined14:44
maveneag_ joined14:44
maveneag_ left14:44
murosai hey, how can i add all files recursively to index that end with .py?14:44
grawity `git add "**/*.py"` if your git version is recent enough14:45
otherwise, `shopt -s globstar; git add **/*.py` if you use bash14:45
Noldorin joined14:45
macben joined14:45
grawity (not sure what the zsh equivalent is)14:45
murosai yup it worked14:45
grawity otherwise, `find -name "*.py" -exec git add {} +`14:46
ni__ left14:46
AaronMT joined14:47
tma996 joined14:47
WAAQMXZ joined14:48
Actreka joined14:48
arietis left14:49
wakeup left14:49
FiLoSsOfIe ce fac puturosiiiiii ?14:49
soee joined14:49
Actreka I was renaming every file in my git-repo however I deleted the files then recreated them (with the new name). Now I can't see the full blame or log. Is there anything I can do to get it back?14:50
petey joined14:50
zorzar joined14:50
psyl0n left14:50
aleng_ joined14:51
karupanerurazz_karupanerura14:51
FerreusR left14:52
FerreusR joined14:52
Kayra joined14:53
aleng_ left14:53
papegaaij left14:53
msm joined14:53
ni__ joined14:55
kcj left14:56
joaoh82 joined14:57
grawity `git blame` follows renames by default, for `git log` use the --follow option14:58
Actreka grawity: The problem is that it doesn't because it was technically not a rename, but a delete and creation14:59
grawity Actreka: that's exactly what a rename is in Git14:59
`git mv` does nothing more.14:59
h_kon left14:59
donmichelangelo left15:00
donmichelangelo joined15:00
LucasKA joined15:01
Luke-Jr left15:01
matanya left15:01
soee left15:01
Actreka grawity: Well, let me explain. This was a while ago and I was not yet used to git. So i deleted all files, then commited, added the renamed files and commited.15:01
joaoh82 left15:01
grawity so it might be that rename-following breaks because there's a commit that lacks those files completely15:02
Luke-Jr joined15:02
Nibato joined15:03
Luke-Jr left15:03
Luke-Jr joined15:03
Actreka Yep, do you know of any way to fix this?15:04
Kayra left15:04
cjwilburn joined15:04
nilg left15:04
LucasKA left15:05
23LAAS91Hadam1215:05
CheckDavid joined15:05
bhldev__ joined15:06
bhldev_ left15:07
Kayra joined15:07
psoo_ left15:07
acharlieh joined15:07
shawndumas joined15:08
mdhopkins joined15:08
dobalina joined15:08
jarr0dsz joined15:09
Psi-Jack left15:10
jarr0dsz left15:10
cff Is anyone using Waf build system in a large project?15:12
stodan left15:12
goTAN joined15:12
joaoh82 joined15:13
looper left15:13
NBeeuwsaert_ joined15:13
ojacobson joined15:15
mklappstuhl joined15:15
NBeeuwsaert left15:16
floatingpoint left15:16
bamj0rb joined15:16
gitinfo set mode: +v15:17
JakePee joined15:17
gain joined15:20
johnkevinmbasco left15:20
becks__ joined15:23
mikecmpbll joined15:24
TRomeo joined15:25
TomyWork left15:25
tang^ joined15:25
TomyWork joined15:26
infynyxx joined15:27
m-o-s joined15:27
h0bbit_ joined15:27
maroloccio joined15:28
LuxuryMode joined15:28
m-o-s_ joined15:28
Macaveli left15:29
mizu_no_oto joined15:30
nw8l joined15:30
ekoekoekoeko joined15:30
mklappstuhl left15:30
petey left15:32
arvind_khadri left15:32
john2x left15:32
koda joined15:32
petey joined15:32
grawity well, Samba uses Waf15:32
m-o-s left15:32
grawity and the project isn't exactly small15:32
psyl0n joined15:34
thevishy Waf is a version control?15:34
oh its a deployment tool15:35
build tool*15:35
tplaner joined15:35
FiLoSsOfIeduc15:36
ducLoRdToLsToI15:36
vtunka joined15:37
petey left15:37
kermit left15:39
pen left15:40
cjsarette left15:41
hits joined15:41
zapb joined15:42
hits left15:43
petey joined15:44
Norm_ left15:44
mudmaste_ joined15:44
Nevik <thevishy> Waf is a version control? <--- best question15:44
tplaner left15:45
tplaner joined15:45
jceb left15:45
Nevik i vote "wtf is version control [asked in #git]" to be the question of the year15:45
strictlyb left15:45
strictlyb joined15:45
notjohn left15:46
dobalina left15:46
dobalina joined15:46
cjsarette joined15:47
jonatasnona joined15:47
thevishy :->15:48
jonatasnona left15:48
jonatasnona joined15:48
mudmaster left15:48
takamichi left15:48
petey left15:50
petey joined15:50
takamichi joined15:51
petey left15:51
petey joined15:51
dfanjul joined15:51
Timmy joined15:51
thierryp left15:52
offby1 "f***ing version control ... how does it work"15:53
Nevik offby1: "how does it work" is a more legit question than "what is a version control", for this channel at least :P15:54
BinGOs not to be conflated with vermin control15:55
Nevik lulz15:55
jargon left15:56
PerlJam "control" is probably too strong of a word anyway. It's more like herding cats than driving a train.15:56
aberrant joined15:56
leandrosansilva_ joined15:57
leandrosansilva left15:57
rasto left15:57
jcvfen joined15:57
nw8l left15:59
shawndumas left15:59
shawndumas joined16:00
tang^ feels like driving a train, given all these "train tracks" git log shows for all these merges16:00
dxdt joined16:00
M00sL0gB0t left16:01
M00sL0gB0t joined16:01
LucasKA joined16:01
Error404NotFound joined16:01
shwaiil joined16:02
shwaiil Hi16:02
kevwil joined16:02
ilteris left16:02
h0bbit_ left16:03
Anticom left16:03
bwreilly joined16:03
shwaiil Q: I've setup a remote repository (using --bare). My local environment now is pushing this to ssh://mydomain.com/var/www/foo, I didn't specified a username to access ssh, forgot, and it seems to work. How is this possible ? Thanks!16:03
DaisyTheCow left16:03
m-o-s_ left16:04
ojacobson shwaiil: same way that 'ssh mydomain.com' works without specifying a username16:04
ssh(1) assumes you meant to use $USER as your username if you don't specify, either in the SSH address or in your ssh config16:04
dougsland left16:04
shwaiil ojacobson: thanks for looking! oh didn't knew that, why is that ?16:04
grawity for convenience, probably16:05
ojacobson (and, of course, you can specify a username to use for mydomain.com in your SSH config, at which point you don't need to specify it in the SSH command or, for git, in the URL)16:05
swaechter left16:06
ni__ left16:06
LucasKA left16:06
dort joined16:06
RaceCondition left16:08
FortySix_ joined16:08
juril left16:08
thevishy left16:08
RaceCondition joined16:09
ThomasLocke left16:09
FortySix2 left16:10
breakingthings left16:11
breakingthings joined16:11
jooools left16:12
lavaman joined16:12
infynyxx left16:12
cbreak-work left16:12
koda left16:12
paulweb515 left16:12
jooools joined16:13
hyperair left16:14
ewet left16:15
jjr joined16:16
senj joined16:16
jjr left16:17
macben left16:18
koda joined16:18
macben joined16:19
ngladitz joined16:19
whiskers75_ joined16:19
whiskers75 left16:19
whiskers75_whiskers7516:19
whiskers75Guest6994516:20
dougsland joined16:20
Error404NotFound left16:20
dougsland left16:20
dougsland joined16:20
rouf_ left16:20
gdogrady joined16:21
master_j joined16:21
Guest69945 left16:21
beyondcreed joined16:21
zapb left16:22
whiskers75_ joined16:22
whiskers75_whiskers7516:23
mlechner left16:23
msm left16:24
St_Marx joined16:25
St_Marx left16:25
axl_ joined16:25
St_Marx joined16:26
joelm joined16:27
petey left16:27
h0bbit_ joined16:28
whiskers75 left16:28
Advocation left16:28
whiskers75 joined16:28
Advocation joined16:29
johnmilton joined16:29
kodab joined16:30
pikaren left16:30
HelperW_ left16:30
petey joined16:30
HelperW_ joined16:31
jnavila left16:31
maestrojed joined16:32
dobalina left16:32
shadowvice joined16:32
bvi left16:32
harobed left16:33
ehmatthes joined16:33
goTAN left16:33
gua_ould left16:34
koda left16:34
kodab left16:35
HelperW_ left16:35
bitslayer joined16:35
infynyxx_ joined16:35
zapb joined16:35
FerreusR left16:35
FerreusR joined16:36
primitive joined16:37
DoniRS left16:38
DolpheenDream_ joined16:38
macben left16:39
Nizumzen left16:39
macben joined16:39
takamichi left16:39
lite_ left16:40
MattMaker left16:41
JDT joined16:41
cagedwisdom left16:41
takamichi joined16:42
macben left16:42
macben joined16:42
YoungFrog left16:43
maroloccio2 joined16:43
Timmy left16:44
ekoekoekoeko left16:45
rath left16:45
maroloccio left16:46
deepwater joined16:46
irqq left16:46
ekoekoekoeko joined16:46
macben left16:47
ColKurtz joined16:47
ColKurtz left16:47
deepwater left16:47
ColKurtz joined16:48
ColKurtz left16:48
mvelioglu left16:48
JEEB left16:48
zackiv31 joined16:48
yhager joined16:48
ColKurtz joined16:48
ColKurtz left16:49
ColKurtz joined16:49
ColKurtz left16:49
ColKurtz joined16:50
ColKurtz left16:50
JDT left16:51
m0viefreak joined16:51
Sonderblade left16:52
zackiv31 left16:53
whiskers75 left16:53
senj left16:53
ColKurtz joined16:53
ColKurtz left16:53
jackneill joined16:54
bwreilly left16:54
shinnya left16:54
jdav_gone left16:54
Luke-Jr left16:54
ColKurtz joined16:54
ColKurtz left16:54
vtunka left16:55
ColKurtz joined16:55
senj joined16:55
noname001__ left16:55
hyperair joined16:55
psyl0n left16:55
whiskers75 joined16:56
Kayra left16:57
SocalNick joined16:57
Timmy joined16:57
RofricTheBearded left16:57
ehmatthes left16:57
thierryp joined16:58
vtunka joined16:58
JakePee left16:58
zackiv31 joined16:58
TRomeo left16:59
shwaiil Q: About http://gitolite.com/deploy.html, what exactly is happening here ( GIT_WORK_TREE=/deploy/dir git checkout -f master ) ? I've read a few times and the reason is to reset the state ?!?16:59
maroloccio2 left16:59
dort left17:00
ni__ joined17:00
diegoviola left17:00
Splenetic joined17:01
JakePee joined17:01
nemesit|znc left17:02
LucasKA joined17:02
ni__ left17:03
ni__ joined17:03
shawndumas left17:03
btree joined17:03
yhager left17:04
paulweb515 joined17:04
Kayra joined17:05
codezomb joined17:05
blinkdesign left17:05
shawndumas joined17:05
takamichi left17:06
roland___ left17:06
ldlework If my local branch is ahead of the origin's branch, how can I see what changes would be pushed?17:07
LucasKA left17:07
hwrdprkns joined17:07
King_Hual joined17:07
tma996 left17:07
luto ldlework: !lol is nice for that17:07
gitinfo ldlework: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all17:07
grawity git fetch origin && git log --oneline @{upstream}..HEAD (or origin/master..master for example)17:07
luto and that fetch thingy to update your remote branches. right.17:08
osse ldlework: the range origin/branch..branch is the commits that would be pushed17:08
kermit joined17:08
osse luto: so git log origin/branch..branch for example17:08
oops17:08
wrong guy17:08
luto osse: I know how to do that, thank you :)17:08
;)17:08
zackiv31 left17:08
m0ltar left17:09
rafeca left17:10
leandrosansilva_ left17:10
FatDarrel joined17:10
leandrosansilva joined17:10
lavong left17:11
pheaver joined17:12
Advocation left17:12
cjsarette left17:12
mitz left17:13
cjsarette joined17:15
Advocation joined17:15
rurufufuss left17:15
osmosis joined17:15
rednaks joined17:16
darnir joined17:16
infynyxx_ left17:16
darnir left17:17
darnir joined17:17
Skrypter joined17:18
bamj0rb shwaiil: GIT_WORK_TREE is an environment variable that overrides the working tree location. Checkout copies the trees and blobs out of the repository and writes them to the working directory as a file system structure. The -f flag tells it to forcibly overwrite changed files (instead of refusing to work).17:18
zackiv31 joined17:18
darnir left17:18
bwreilly joined17:18
darnir joined17:18
breakingthings I have a file with changes that have not been committed and I want to blame the file but without the changes in the working directory (to see when the change I'm fixing happened). Is there an easy way to do this other than stashing those changes17:19
yhager joined17:19
ojacobson 'git blame HEAD -- <paths>'17:20
imachuchu joined17:20
JEEB joined17:20
maroloccio2 joined17:21
breakingthings thanks17:21
croaky joined17:22
jefferai joined17:22
ChanServ set mode: +o17:23
senj left17:23
suy_ joined17:23
livingstn joined17:23
CalimeroTeknik left17:23
EugeneKay changed the topic to: Welcome to #git, the place for git help and blood sausage | Current stable version: 1.8.5.2 | Start here: http://jk.gs/git | Seeing "Cannot send to channel" or unable to change nick? /msg gitinfo .voice | Keep good care of your git-shell and your inner snail will thank you17:23
EugeneKay set mode: +v17:23
EugeneKay set mode: -o17:23
jargan joined17:23
looper joined17:23
suy left17:24
jefferai_gone left17:24
jast left17:24
jarganjast17:24
mar77i joined17:24
mikecmpbll left17:24
obiwahn left17:24
mikecmpbll joined17:24
eaj left17:26
mar77i hm. there is core.whitespace tabwidth=<n> in the manpage, although all online sources suggest use of setting core.pager to less -x <n>. is this the suggested way to do this or can tabwidth be used for this?17:26
mitz joined17:26
obiwahn joined17:26
cdleonard1 joined17:27
PLejeck joined17:27
eaj joined17:27
mar77i to answer my question right away, core.whitespace seems to concern post-processing at commits, right?17:27
bitslayer left17:27
konadr left17:28
msafi left17:28
CalimeroTeknik joined17:29
CheckDavid left17:29
akiva joined17:29
mizu_no_oto left17:30
Sp4rKy_ joined17:30
msafi joined17:30
jdav_gone joined17:30
syst3mw0rm_ joined17:30
donmichelangelo left17:30
canton7 mar77i, core.whitespace affects colored 'git diff' output and 'git apply' operation17:30
(as stated by man git-config)17:30
bartek left17:30
donmichelangelo joined17:30
bartek joined17:31
infynyxx joined17:31
bartek left17:31
bartek joined17:31
FerreusR left17:31
ojacobson mar77i: it's worth noting that (a) there's no widespread convention for passing whitespace settings to external tools (like pagers), and (b) external tools are not expected to look at git-config values themselves17:31
daito joined17:32
PLejeck left17:32
PLejeck joined17:32
gdogrady left17:32
mikecmpbll left17:32
tang^ left17:33
Actreka left17:33
eaj left17:33
tang^ joined17:33
ekoekoekoeko left17:34
Xiti left17:34
cdleonard left17:34
nuck left17:34
russ__ left17:34
D4N3 left17:34
Sp4rKy left17:34
Aaeriele left17:34
syst3mw0rm left17:34
DrZaius_ left17:34
whoami left17:34
BeerSerc left17:34
stkrzysiak left17:34
Silmarilion joined17:34
wkn_Mako left17:34
Xiti joined17:34
whoami joined17:34
Timmy left17:35
c0ded left17:35
stkrzysiak joined17:35
russ__ joined17:35
psyl0n joined17:35
PLejecknuck17:35
freimatz left17:35
GameGamer43 left17:35
GameGamer43_ joined17:35
gitinfo the git-config manpage is available at http://jk.gs/git-config.html17:35
ojacobson gitinfo's having a rough day :)17:36
unkmar_ left17:36
shruggar left17:36
cjsarette left17:36
daito left17:36
alorence left17:37
cjsarette joined17:37
c0ded joined17:38
vince_prignano joined17:38
ekoekoekoeko joined17:39
Nizumzen joined17:40
GameGamer43_GameGamer4317:40
boombatower joined17:41
VictorCL joined17:41
joelm_ joined17:42
Aaeriele joined17:42
bitkiller_ left17:42
VictorCL left17:43
DrZaius_ joined17:43
JosephFerano_ joined17:43
mudmaster joined17:43
gr_wity joined17:43
becks___ joined17:43
bitkiller_ joined17:43
bwreilly_ joined17:43
kapowaz_ joined17:43
alezost` joined17:43
grawity left17:43
gr_witygrawity17:43
hjp_beta__ joined17:44
dwmw2_go` joined17:44
PebkacJones joined17:44
PebkacJones hello, gits17:45
FortySix2 joined17:45
FatDarrel_ joined17:45
konadr joined17:46
mitz_ joined17:46
adambrenecki_ joined17:46
maroloccio2 left17:46
smef left17:46
WillPittenger joined17:46
mikaelj_ joined17:46
kevwil left17:46
dsop joined17:46
stealthii_ joined17:47
PebkacJones left17:47
EugeneKay You're a git17:47
harobed_ joined17:47
ryanakca_ joined17:47
harobed_ left17:47
RaceCondition left17:47
Forkk13 joined17:47
wcpan_ joined17:47
nickoe__ joined17:47
edin- joined17:47
titusjose__ joined17:47
harobed_ joined17:48
ELLIOTTCABLE_ joined17:48
Nevik gitty in here, i'nt it17:48
harobed_ left17:48
WillPittenger I'm trying to clone a URL from a private GitHub project. But the URL given by GitHub, [email@hidden.address] isn't working. Putty pops up with an error.17:48
krkini joined17:48
_Parnic joined17:48
WillPittenger Disconnected: No supported authentication methods available.17:48
B:\>git clone -v --progress --recursive "git@github.com:doctorleff/u3e.git" b:\tempCloning into 'b:\temp'...fatal: Could not read from remote repository.17:48
zumba_addict hey folks, i just finished running git pull from the remote main-bug-fixes-branch. I have the latest codes now. When I create a new local branch, does it mean that the code will be the same as main-bug-fixes-branch? I'm confused because my brain is telling me that when I create a new branch, it will be the same as master and not main-bug-fixes-branch.17:48
mvelioglu joined17:48
WillPittenger That's what shows up on the command line.17:48
WhereIsMySpoon joined17:49
m0viefreak WillPittenger: do you have ssh access to that repo?17:49
dfarler_ joined17:49
tobias__ joined17:49
WillPittenger I'm a project member. I don't know about the rest.17:49
WhereIsMySpoon Hi, in git extensions it gives you a nice ui of the stuff that you changed since your last commit on the branch you're on. How do I do this via git on the cmd line?17:49
bwreilly_ left17:49
zumba_addict main-bug-fixes-branch is another separate branch that is being used on another dev server.17:49
EugeneKay WillPittenger - you need to set up a SSH key with your Github account....17:49
zumba_addict what are your thoughts folks?17:49
shihan_ joined17:50
hahuang65 left17:50
WhereIsMySpoon also why is #git the place for blood sausage? O.o17:50
EugeneKay WillPittenger - and your client(putty on windows) may need to have the key loaded into memory if it's passphrase protected(Pageant)17:50
ddv_ joined17:50
akitada_ joined17:50
EugeneKay Or, use a https:// URL17:50
harobed_ joined17:50
konadr left17:50
ngladitz_ joined17:50
harobed_ left17:50
dytrivedi_ joined17:50
cjsarette left17:50
shihan left17:50
kapowaz left17:50
netiul left17:50
hjp_beta_ left17:50
steverandy left17:50
JosephFerano left17:50
JosephFerano_JosephFerano17:50
dwmw2_gone left17:50
dytrivedi left17:50
knownasilya left17:50
anarchih left17:50
troyt left17:50
offbyone left17:50
bwreilly left17:50
tapout left17:50
Parnic left17:50
Novacha left17:50
thunderbolt left17:50
GameGamer43 left17:50
zapb_ joined17:50
zapb left17:50
joelm left17:50
St_Marx left17:50
mudmaste_ left17:50
mikaelj left17:50
dmarr left17:50
Kyle left17:50
c0ded left17:50
mitz left17:50
FatDarrel left17:50
paulweb515 left17:50
ngladitz left17:50
FortySix_ left17:50
solrize left17:50
FatDarrel_FatDarrel17:50
simpleminds left17:50
kini left17:50
akitada left17:50
dsop_ left17:50
alezost left17:50
salamanderrake left17:50
cwillu_at_work left17:50
DANtheBEASTman left17:50
gnarf left17:50
rastus_vernon left17:50
Edgan left17:50
aspiers left17:50
bedouin left17:50
Celelibi left17:50
neozonz left17:51
WillPittenger I get this when I do that:17:51
ozymandias left17:51
ryanakca left17:51
Kamilion left17:51
emmajane left17:51
whaley left17:51
ConstantineXVI left17:51
adambrenecki left17:51
dfarler left17:51
Forkk left17:51
dfarler_dfarler17:51
Mez left17:51
stealthii left17:51
nickoe left17:51
crane left17:51
titusjose_ left17:51
ddv left17:51
jgornick left17:51
ELLIOTTCABLE left17:51
edin left17:51
atrus left17:51
wcpan left17:51
dan64 left17:51
becks__ left17:51
thunderbolt__ joined17:51
dmarr joined17:51
simpleminds joined17:51
Guest41679 joined17:51
Guest41679 left17:51
Guest41679 joined17:51
ozymandias joined17:51
whaley joined17:51
Celelibi joined17:51
GameGamer43_ joined17:51
kapowaz_kapowaz17:51
GameGamer43_ left17:51
GameGamer43_ joined17:51
WillPittenger B:\>git clone -v --progress --recursive "https://github.com/doctorleff/u3e.git"17:51
alezost`alezost17:51
WillPittenger b:\temp17:51
Cloning into 'b:\temp'...17:51
fatal: could not read Username for 'https://github.com': No such file or directory17:51
DANtheBEASTman joined17:51
dytrivedi_dytrivedi17:51
Novacha joined17:51
strictlybsb17:51
Edgan joined17:51
aspiers_ joined17:51
m0viefreak you can't use https with private repos i think17:51
Kyle joined17:51
mutex_mutex17:51
paulweb515 joined17:51
titusjose__titusjose_17:51
jgornick joined17:51
EugeneKay You sure can; just need to specify user@17:51
grawity you can, as long as you authenticate17:51
ddv_ddv17:51
salamanderrake joined17:51
solrize joined17:51
KyleGuest2128717:51
bubbleheadinc joined17:51
EugeneKay man git-clone for how to set up the URL spec17:51
neozonz joined17:51
cwillu_at_work joined17:51
ozymandias left17:51
ozymandias joined17:51
paulweb515 left17:52
paulweb515 joined17:52
gnarf joined17:52
Novacha left17:52
Novacha joined17:52
netiul joined17:52
solrize left17:52
solrize joined17:52
neozonz left17:52
neozonz joined17:52
neozonz left17:52
neozonz joined17:52
stealthii_stealthii17:52
WillPittenger EugeneKay: There is no man on Windows.17:52
msafi left17:52
adambrenecki_adambrenecki17:52
steverandy joined17:52
Gamecubic joined17:52
EugeneKay Normally gitinfo provides links17:52
netiul left17:52
netiul joined17:52
netiul left17:52
netiul joined17:52
atrus joined17:52
EugeneKay pokes it with a stick17:52
Mez joined17:52
knownasilya joined17:52
crane joined17:52
ELLIOTTCABLE_ELLIOTTCABLE17:52
anarchih joined17:52
c0ded joined17:52
akitada_akitada17:52
bremner WillPittenger: not even in git bash17:52
dan64 joined17:52
bremner ?17:52
bedouin joined17:52
St_Marx joined17:52
Gamecubic left17:52
WhereIsMySpoon bremner: not in mine17:52
c0ded left17:52
c0ded joined17:52
Mez left17:53
Mez joined17:53
bremner you need the man pages. it's not negotiable.17:53
WhereIsMySpoon theyr eonline anyway17:53
bremner sure.17:53
Nevik git for windows provides man pages in html format17:53
cjsarette joined17:53
Nevik running git --help command or git command --help launches them17:53
Gamecubic joined17:53
WillPittenger bremner: git help <command> opens webpages that look like man.17:53
gitinfo the git-clone manpage is available at http://jk.gs/git-clone.html17:53
osse On linux git help <command> launches man git-command :)17:53
Nevik git for windows does not bring textmode manpages17:53
bremner then you have man pages. H.A.N.D.17:53
[emmajane] joined17:53
jerichodotm left17:54
EugeneKay There it is. !botsnack17:54
gitinfo Om nom nom17:54
jerichodotm joined17:54
WhereIsMySpoon anyone got anything for my q?17:54
koda joined17:54
CocoStorm joined17:54
jerichodotm left17:54
jerichodotm joined17:54
vals_ joined17:54
WhereIsMySpoon i said it above but lots of people suddenly reconnected lol >.<17:54
CocoStorm how do I delete a contributor from my repo?17:55
hobodave joined17:55
EugeneKay Either !rewrite them out of history by hiding the !sausage or..... don't.17:55
gitinfo [!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum17:55
[!sausage_making] Some developers like to "hide the sausage making", transforming their commits before presenting them to the outside world. See http://sethrobertson.github.com/GitBestPractices/#sausage and !perfect17:55
ConstantineXVI joined17:55
tango_ left17:55
a-l-e joined17:55
xrq WhereIsMySpoon: what exactly does it show you?17:55
Kayra left17:55
moritz CocoStorm: what exactly do you want to do? Remove access for a user?17:55
tapout joined17:56
vals_tango_17:56
Kamilion joined17:56
troyt joined17:56
CocoStorm moritz, yeah, so said person can no longer see my code17:56
bremner I bet 5 cents he means from github17:56
CocoStorm haha17:56
yes..17:56
xrq WhereIsMySpoon: it doesn't show you something like git log --oneline --graph --decorate --all, does it?17:56
CocoStorm from a private repo17:56
harobed_ joined17:56
WhereIsMySpoon__ joined17:56
harobed_ left17:56
moritz CocoStorm: where are you hosting that repo? github?17:56
WhereIsMySpoon__ xrq: i dont know what to do to get it to show the changed files since last commit17:56
CocoStorm moritz, yes17:56
xrq WhereIsMySpoon__: git diff --names-only17:56
WhereIsMySpoon left17:57
WhereIsMySpoon__WhereIsMySpoon17:57
WhereIsMySpoon left17:57
WhereIsMySpoon joined17:57
WhereIsMySpoon left17:57
WhereIsMySpoon joined17:57
jnavila joined17:57
harobed_ joined17:57
Sly1986 left17:57
harobed_ left17:57
moritz CocoStorm: Settings (on the right) -> Collaborators17:57
EugeneKay Oh, well that's easy. Go into the settings icon and remove them from the Contributors page17:57
xrq *name-only17:57
WhereIsMySpoon xrq: error: invalid options --names-only17:57
ah17:57
EugeneKay Same place you added them17:57
moritz CocoStorm: each contributor has a read (remove) link17:58
harobed_ joined17:58
harobed_ left17:58
WhereIsMySpoon xrq: that doesnt seem to be obeying my gitignore17:58
xrq: ie it lists .obj files amongst others when .obj is in my .gitignore17:58
harobed_ joined17:58
CocoStorm ahh I see moritz it's quite hard to find :P17:59
harobed_ left17:59
CocoStorm heh found it in the end thanks!17:59
WhereIsMySpoon !sausage17:59
gitinfo [!sausage_making] Some developers like to "hide the sausage making", transforming their commits before presenting them to the outside world. See http://sethrobertson.github.com/GitBestPractices/#sausage and !perfect17:59
harobed_ joined17:59
marhaban left17:59
harobed_ left18:00
zumba_addict anyone please?18:00
Bass10 joined18:00
harobed_ joined18:00
b3tamax left18:00
harobed_ left18:00
WhereIsMySpoon xrq: how can I make it take into account my gitignore?..18:01
xrq WhereIsMySpoon: sounds like a good bug to file18:01
koda left18:01
FerreusR joined18:01
zumba_addict or does git refers to the current branch as the source when you make a new local branch?18:01
harobed_ joined18:01
harobed_ left18:01
dreiss joined18:01
moritz zumba_addict: it does18:01
zumba_addict: unless you explicitly give it another base18:02
zumba_addict ok18:02
i wasn't aware of that. I thought it will always refer to master18:02
nickoe__nickoe18:02
corni joined18:02
matanya joined18:02
CocoStorm left18:02
zumba_addict currently, I'm on main-bug-fixes-branch branch. That means, when I create a new local branch, the code willb e the same as main-bug-fixes-branch. Correct?18:02
osse zumba_addict: by default yes. but you can explicitly name a commit to start the new branch on if you wish18:03
LucasKA joined18:03
WhereIsMySpoon xrq: id rather assert that it isnt my own idiocy making it happen by me not setting up soimething rather than automatically saying its a bug18:03
hobodave_ joined18:03
zumba_addict got it osse. Another question18:03
FortySix2 left18:04
koda joined18:04
ColKurtz left18:04
sami_ joined18:04
zumba_addict I did a git pull on main-bug-fixes-branch. I have an existing local branch called css-bug-fix which I created yesterday. How do I update my css-bug-fix so that it can have the new changes that my main-bug-fixes-branch received few mins ago?18:05
dxdt left18:05
gdogrady joined18:05
osse zumba_addict: git checkout css-bug-fix; git merge main-bux-fixes-branch18:05
ahmedelgabri left18:06
FerreusR left18:06
zumba_addict cool18:06
doing it now18:06
dinoangelov left18:06
harobed_ joined18:06
harobed_ left18:06
peteypetey_18:06
zumba_addict our, I have conflicts18:06
ouch18:07
ChrisAnn left18:07
harobed_ joined18:07
moritz !eek18:07
gitinfo [!eekaconflict] Merge conflicts are a natural part of collaboration. When facing one, *don't panic*. Read "How to resolve conflicts" in man git-merge and http://git-scm.com/book/ch3-2.html#Basic-Merge-Conflicts then carefully go through the conflicts. Picking one side verbatim is not always the right choice!18:07
maroloccio joined18:07
zumba_addict ok18:07
harobed_ left18:07
ChrisAnn joined18:07
aspiers_aspiers18:07
harobed_ joined18:08
LucasKA left18:08
harobed_ left18:08
dprophit joined18:09
Mez_ joined18:09
harobed_ joined18:09
zumba_addict that means I'll have to check the codes from <<<<<<<< until >>>>>>>>18:09
harobed_ left18:09
WhereIsMySpoon left18:09
tapout left18:09
jgornick left18:09
Edgan left18:09
Guest41679 left18:09
ConstantineXVI left18:09
Kamilion left18:09
troyt left18:09
[emmajane] left18:09
salamanderrake left18:09
Novacha left18:09
ozymandias left18:09
hobodave left18:09
St_Marx left18:09
dan64 left18:09
c0ded left18:09
solrize left18:09
atrus left18:09
cwillu_at_work left18:09
gnarf left18:09
DANtheBEASTman left18:09
GameGamer43_ left18:09
Guest21287 left18:09
whaley left18:09
dmarr left18:09
simpleminds left18:09
Mez left18:09
Celelibi left18:09
Edgan joined18:09
dan64 joined18:09
Novacha joined18:09
GameGamer43_ joined18:09
ozymandias joined18:09
Guest41679 joined18:09
DANtheBEASTman joined18:10
zumba_addict I'm assuming mine is <<<<<<<<HEAD18:10
ekoekoekoeko left18:10
jgornick joined18:10
Celelibi joined18:10
solrize joined18:10
[emmajane] joined18:10
cwillu_at_work joined18:10
salamanderrake joined18:10
ozymandias left18:11
ozymandias joined18:11
shwaiil left18:11
solrize left18:11
solrize joined18:11
gnarf joined18:11
psyl0n left18:11
ekoekoekoeko joined18:11
simpleminds joined18:11
ColKurtz joined18:11
gdogrady left18:11
Novacha left18:11
Novacha joined18:11
Norm_ joined18:11
whaley joined18:11
atrus joined18:11
tapout joined18:11
ConstantineXVI joined18:11
Kamilion joined18:11
dmarr joined18:11
troyt joined18:11
Kyle- joined18:11
c0ded joined18:11
[emmajane] left18:11
koda left18:12
h0bbit_ left18:12
harobed_ joined18:12
harobed_ left18:12
Edgan Is there any way to have git not change files on conflict, by default? I don't like that the default is leave me in a broken state on conflict.18:13
harobed_ joined18:14
petey_ left18:14
St_Marx joined18:14
harobed_ left18:14
petey_ joined18:14
harobed_ joined18:15
harobed_ left18:15
harobed_ joined18:16
harobed_ left18:16
[emmajane] joined18:16
jrnieder Edgan: which version of the file do you want to show up on conflict?18:16
grawity I think the problem is that it has to start doing the actual merge until it can find out that a file conflicts18:16
harobed_ joined18:17
chrisss404 left18:17
grawity though, `git pull --ff-only` *may* be somewhat useful18:17
jrnieder ah, a sort of "merge --dry-run" before the real merge. yeah, that would be nice18:17
harobed_ left18:17
grawity (but that rejects all merges, not merely conflicting ones)18:17
jrnieder charon wrote an index-only merge that might be a nice building block18:17
harobed_ joined18:17
harobed_ left18:17
jrnieder http://thread.gmane.org/gmane.comp.version-control.git/22978718:18
harobed_ joined18:19
Edgan jrnieder: What ever there was before. From what I just read I want git merge --abort on "automatic"18:19
harobed_ left18:19
petey_ left18:19
jrnieder ok, sounds like you want what grawity is describing18:19
maroloccio left18:19
geomyidae left18:19
harobed_ joined18:19
harobed_ left18:19
mklappstuhl joined18:20
DrShoggoth joined18:20
Edgan yeah --ff-only sounds useful18:20
koda joined18:20
VictorCL joined18:20
harobed_ joined18:20
harobed_ left18:20
Targen left18:20
Edgan grawity: jrnieder: thanks18:21
grawity yeah, if you're doing it from any kind of a "deploy" script, --ff-only is basically a requirement18:21
danman joined18:21
mrhooray joined18:21
grawity since you usually wouldn't want any merge to happen at all18:21
ekoekoekoeko left18:21
harobed_ joined18:21
harobed_ left18:21
irqq joined18:21
xlinkz0 left18:22
SocalNick left18:22
harobed_ joined18:22
Nizumzen left18:22
harobed_ left18:22
geomyidae joined18:23
thevishy joined18:23
harobed_ joined18:23
harobed_ left18:23
keeb left18:24
harobed_ joined18:24
harobed_ left18:24
Advocation left18:24
SocalNick joined18:24
FerreusR joined18:25
mizu_no_oto joined18:25
CheckDavid joined18:25
hahuang65 joined18:25
doev left18:26
kodab joined18:26
harobed_ joined18:27
doev joined18:27
harobed_ left18:27
Mars___ joined18:28
leeN joined18:28
Timmy joined18:28
koda left18:29
geomyidae left18:29
Baluse left18:30
kukks joined18:31
zumba_addict i've fixed the conflicts by editing them in vi. When i run git merge source-branch, it's saying "merge is not possible because you have unmerged files" I don't see the conflict files anymore. Am I missing a command?18:31
jrnieder zumba_addict: 'git status'18:32
zumba_addict i see the files I modified18:32
do I have to commit them?18:32
jrnieder yup18:32
zumba_addict cool18:32
b3tamax joined18:32
bremner you have to add them18:32
zumba_addict didn't know that :)18:32
will do18:32
mrhanky joined18:32
bremner and commit them too, sorry. Brain vacation.18:33
sami_ left18:33
zumba_addict perfect, it's good now. Merged :D18:33
thanks folks18:33
mrhanky do unix sym-links in a git repo work under windows?18:33
geomyidae joined18:33
harobed_ joined18:34
rednaks left18:34
geomyidae left18:34
mrhanky example: <repo>/foo.ext to <repo>/bar/foo.ext18:34
jrnieder mrhanky: I think it tries to gracefully degrade18:34
grawity barely – msysgit will check out them as regular files, IIRC (it doesn't know how to create Windows symlinks yet)18:34
mrhanky degrade?18:34
grawity meanwhile, Cygwin git will create Cygwin-compatible symlinks which don't work with native Win32 programs18:34
jrnieder what grawity said18:34
Windows symlinks = reparse points?18:35
mrhanky so git on windows will copy the linked file to destination?18:35
grawity jrnieder: a type of reparse points, yes18:35
jrnieder mrhanky: try it? :)18:35
WillPittenger Problem and conflict fixed. Thanks guys.18:36
WillPittenger left18:36
m0viefreak http://cygwin.com/cygwin-ug-net/using-cygwinenv.html18:36
winsymlinks:native uses NTFS reparse points18:36
jrnieder mrhanky: I think it will just create a text file naming the symlink target18:36
mrhanky i dont have a windows system jrnieder :)18:36
grawity jrnieder: right (which is how symlinks are stored internally too)18:36
it's a bit unfortunate that CreateSymbolicLink() requires privileges :(18:36
pheaver left18:36
aChatir joined18:37
przemoc left18:39
petey_ joined18:39
cjsarette left18:40
aChatir left18:40
przemoc joined18:40
geomyidae joined18:40
cjsarette joined18:40
tirengarfio left18:42
jhulten joined18:43
mf_ left18:44
msafi joined18:44
vurma left18:44
Gamecubic left18:44
aard_ joined18:44
tang^ jrnieder: windows junctions are roughly equivalent to symlinks18:44
mf_ joined18:44
Gamecubic joined18:44
mf_Guest3103218:45
fire left18:45
mr_red left18:45
grawity not quite the same thing though – and Windows *does* have symlinks as of Vista18:46
mr_red joined18:46
msafi left18:47
kodab left18:47
VictorCL left18:47
DrShoggoth left18:47
matanya left18:47
a-l-e left18:47
paulweb515 left18:47
bitkiller_ left18:47
russ__ left18:47
donmichelangelo left18:47
obiwahn left18:47
osmosis left18:47
codezomb left18:47
whiskers75 left18:47
maestrojed left18:47
jooools left18:47
hays_ left18:47
SamB left18:47
Onoz left18:47
Schrostfutz left18:47
Schrostfutz joined18:47
donmichelangelo joined18:47
whiskers75 joined18:47
Onoz joined18:47
msafi joined18:47
russ__ joined18:47
kodab joined18:47
bitkiller_ joined18:48
VictorCL joined18:48
sq-one joined18:48
osmosis joined18:48
paulweb515 joined18:48
maestrojed joined18:48
jooools joined18:48
obiwahn joined18:48
DrShoggoth joined18:48
SamB joined18:48
a-l-e joined18:48
gdogrady joined18:49
yhager left18:49
cjwilburn left18:49
matanya joined18:49
ColKurtz left18:49
LBV_User joined18:49
FortySix2 joined18:50
ncd|leeN joined18:50
Gamecubic_ joined18:50
mrhooray left18:51
mr_red- joined18:52
danman left18:52
joelm_ left18:52
mr_red left18:52
mr_red-mr_red18:52
aard__ joined18:52
joelm joined18:52
deandean|away18:52
valgrind joined18:52
LucasKA joined18:52
ColKurtz joined18:53
tirengarfio joined18:54
ewnd9 left18:54
harobed_ left18:54
dxdt joined18:54
Baluse joined18:55
Hydrant joined18:56
ColoradoPods joined18:57
ColoradoPods I see the calender. But Steve just tweeted that recording starts in 5 minutes18:57
psyl0n joined18:57
LucasKA left18:57
Hydrant hi all, I'm slowly switching to git from svn.... I have a repo that I need to extensively change, and based on notes from a video I should do "git checkout -b topic-name master" and then I just do whatever I want... my understanding is that I can also examine where I branched from (look at files I deleted in the branch)... is that correct?18:58
ncd|leeN left19:00
GameGamer43__ joined19:00
zhenjunyao joined19:00
Freeman42 joined19:00
grawity Hydrant: you could use `git show master:path/to/file`, or various invocations of `git diff` to compare the versions in master & working tree, or master & current branch19:01
Celelibi_ joined19:01
GameGamer43_ left19:01
Edgan left19:01
ConstantineXVI left19:01
troyt left19:01
jgornick left19:01
Novacha left19:01
[emmajane] left19:01
St_Marx left19:01
Guest41679 left19:01
dmarr left19:01
salamanderrake left19:01
DANtheBEASTman left19:01
dan64 left19:01
DrShoggoth left19:01
Onoz left19:01
Gamecubic left19:01
aard_ left19:01
geomyidae left19:01
leeN left19:01
whaley left19:01
Kamilion left19:01
c0ded left19:01
simpleminds left19:01
tapout left19:01
solrize left19:01
atrus left19:01
Celelibi left19:01
Kyle- left19:01
gnarf left19:01
cwillu_at_work left19:01
ozymandias left19:01
croaky_ joined19:01
grawity (or, if you need to do this very often during the "extensive changes", you could clone the repository a second time, and check out master there)19:01
Hydrant grawity: when I do "git checkout topic-name" I can switch around branches, and have changes flow back and forth under?19:02
panpainter joined19:02
Hydrant normally with svn I do two checkouts, but I'm trying to learn the git ways19:02
DrShoggoth joined19:02
Hydrant I basically need to rewrite a bunch of code but I need to see the old stuff for reference19:02
grawity both ways would work19:03
Timmy left19:03
Hydrant when I do git checkout master... does this change the "view" of the files back to what is from the repo?19:03
musicmatze joined19:03
ncthom91 joined19:03
croaky left19:03
grawity well, yes19:03
FortySix2 left19:04
Timmy joined19:04
grawity I think `git checkout foo` is similar to `svn switch ^/branches/foo`19:04
moritz it changes the working copy to branch master.19:04
musicmatze left19:04
canton7 !float19:04
gitinfo If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there. If the files you changed differ between branches, the checkout will fail. In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.19:04
FortySix2 joined19:04
mar77i_ joined19:04
mklappstuhl left19:04
Hydrant so it seems like I should just have two checkouts for this type of thing19:05
b3tamax left19:05
eaj joined19:05
ConstantineXVI joined19:05
PerlJam Hydrant: s/checkout/clone/ I think you mean. If so, that's an option19:05
grawity if you need to compare the files often, then maybe yes19:05
jason237 left19:05
Hydrant okay thanks for the help, trying to get used to git ;-)19:06
jason237 joined19:06
mizu_no_oto left19:06
cwillu_at_work joined19:06
salamanderrake joined19:06
Kyle joined19:06
leeN joined19:06
emmajane joined19:06
Guest41679 joined19:06
gnarf joined19:06
atrus joined19:06
St_Marx joined19:06
Onoz joined19:06
c0ded joined19:06
ozymandias joined19:06
whaley joined19:06
simpleminds joined19:06
solrize joined19:06
DANtheBEASTman joined19:06
dan64 joined19:06
halu joined19:06
jgornick joined19:06
dmarr joined19:06
Novacha joined19:06
Edgan joined19:06
sahil__ joined19:06
valgrind left19:06
croaky joined19:06
c0ded left19:06
DANtheBEASTman left19:06
Onoz left19:06
halu left19:06
Kyle left19:06
DANtheBEASTman joined19:06
Onoz joined19:06
Kyle joined19:06
mar77i left19:07
Novacha left19:07
Novacha joined19:07
ozymandias left19:07
ozymandias joined19:07
c0ded joined19:07
b3tamax joined19:07
elastic_dogelasticdog19:07
przemoc left19:07
c0ded left19:07
c0ded joined19:07
vtunka left19:07
przemoc joined19:07
dan64Guest9543919:07
gnarfGuest1206019:07
KyleGuest1627719:07
przemoc left19:07
przemoc joined19:07
tapout joined19:07
VictorCL left19:07
Kamilion joined19:07
troyt joined19:07
croaky_ left19:07
halu joined19:07
imachuchu Hydrant: I always use difftool (which allows you to diff your current file with a version in a different commit) or you could look into git-fs which is a fuse interface for your git repo19:08
geomyidae joined19:08
valgrind joined19:08
mfulz joined19:09
xlinkz0 joined19:10
darnir left19:10
Hydrant imachuchu: thx19:11
grawity git-fs is probably a little too much... unless you're used to having branches and tags as directories, I guess :D19:11
luto grawity: noooooo >.<19:12
jason237 left19:12
luto grawity: please never ever use git and that fact in the same sentence lol19:12
catsup is there a convenient way to check whether git-checkout would treat an argument as a file or as a branch?19:12
Guest16277 left19:13
Guest16277 joined19:13
Guest16277Kyle19:13
luto parse-ref?19:13
przemoc left19:13
jrnieder catsup: just always include the disambiguating '--'?19:13
Mars___ left19:13
gdogrady left19:14
catsup jrnieder: that won't work for my application19:14
grawity what is your application, then?19:14
jrnieder then luto is right, modulo spelling :)19:14
catsup: http://thread.gmane.org/gmane.comp.version-control.git/238853/focus=238958 may also be relevant to your interests19:14
luto ref-parse?19:14
catsup git-rev-parse ?19:14
jrnieder rev-parse19:14
luto oh.19:14
zumba_addict is there a command which will show the files I have commited?19:14
luto sorry :)19:15
zumba_addict: the files you have commited?19:15
zumba_addict: git log maybe?19:15
zumba_addict yup19:15
k19:15
catsup grawity: wrapping git-checkout with user-supplied arguments19:15
zumba_addict not the one i push19:15
jrnieder catsup: um :)19:16
luto I don't really understand you, sorry..19:16
jrnieder catsup: can you say more about the application?19:16
catsup it's just a git wrapper19:16
przemoc joined19:16
jrnieder that's a big "just" :)19:16
zumba_addict btw, how do I rename my local branch? I have not pushed it yet19:17
ojacobson catsup: can you say what it is about your application that makes explicit disambiguation impossible?19:17
jrnieder makes me worry that I'm going to break your code some day by introducing a new commandline argument or something19:17
ojacobson zumba_addict: man git-branch19:17
gitinfo zumba_addict: the git-branch manpage is available at http://jk.gs/git-branch.html19:17
catsup ojacobson: the user might not supply --19:17
ojacobson catsup: that's true; why is that a problem?19:17
bamj0rb catsup: That is the user's responsibility.19:17
ojacobson catsup: remember, we can't see over your shoulder and have _no idea_ what the interface to your wrapper looks like.19:17
jrnieder what does the wrapper do? is it something that hooks could take care of, if only there were more hooks?19:18
catsup ojacobson: that's ok, you don't have to write it19:18
TomyWork left19:18
joaoh82 left19:18
rafaqueque joined19:18
zumba_addict there are two, -m and -M19:18
bamj0rb Your wrapper should document the same rule. Or else, do its own option parsing and pass that along in a way that Git will understand unambiguously.19:18
jrnieder ah, I might not have understood what wrapper means19:18
doev left19:18
pbrobst joined19:18
shruggar joined19:19
mizu_no_oto joined19:19
jrnieder is it a replacement for 'git checkout' where users would expect to be able to pass arbitrary git-checkout arguments to it?19:19
or something else?19:19
catsup jrnieder: yes19:19
i could actually do a hook for this particular one, but other things i'm wrapping don't have the required hooks19:20
Advocation joined19:20
jrnieder git maintainers really really want to make sure the hooks people need exist19:20
it makes their life down the line easier19:20
catsup hm19:20
pr0ton joined19:21
qsusOFFcsqsuscs19:22
robbyoconnor joined19:23
msafi left19:23
pantsman joined19:23
LucasKA joined19:23
GameGamer43__ left19:24
D4N3 joined19:24
PeterBooker left19:24
w0ng left19:25
GameGamer43 joined19:25
cjwilbur_ joined19:25
pinage404 joined19:25
singingwolfboy joined19:26
SocalNick left19:26
tirengarfio left19:26
singingwolfboy is there a command I can run to non-interactively squash all commits between two sha1 hashes that I specify?19:26
SocalNick joined19:27
mar77i joined19:28
mar77i_ left19:28
LucasKA left19:28
tirengarfio joined19:28
SocalNick left19:28
SocalNick joined19:29
panpainter left19:29
Timmy left19:30
sahil__ left19:31
Mars___ joined19:31
w0ng joined19:32
longbeach joined19:32
_ikke_ Do you want to alter history, or want to create a squash-merge?19:32
singingwolfboy _ikke_: what exactly is a squash-merge?19:33
_ikke_ singingwolfboy: Basically, it merges two branches, but doesn't create a merge commit, just a regular commit19:34
singingwolfboy _ikke_: ah. No, I want to alter history19:35
Mars___ left19:35
kukks left19:35
w0ng left19:35
Guest93929 left19:35
irqq left19:36
tirengarfio left19:36
w0ng joined19:36
_ikke_ afaik, there is not non-interactive way to do that.19:37
s/not/no/19:37
rpg joined19:37
veebers joined19:37
petey__ joined19:38
mar77i_ joined19:38
geekbri_ joined19:38
eaj left19:39
bamj0rb You can set the editor to something that will automagically edit the file for you.19:40
Sometime I do EDITOR='sed ...' git ...19:40
kukks joined19:40
kukks left19:40
kukks joined19:40
petey__ left19:40
petey__ joined19:41
_ikke_ right19:41
petey_ left19:41
shawndumas left19:41
mar77i left19:41
bamj0rb Perhaps something like: EDITOR='sed -e 2,$s/^pick/squash/' git rebase -i <sha1>19:41
blahdodo left19:42
bamj0rb That's more for fun than profit.19:42
geekbri left19:42
cbreak|evilcbreak19:43
Silmarilion left19:43
nilg joined19:44
canton7 most decent editors allow you to edit multiple lines at once19:45
tirengarfio joined19:45
_ikke_ :%s/pick/squash/19:46
But that's still interactive19:46
moritz and you can'T squash the very first line19:47
dev__ left19:47
Advocation left19:47
moritz I wonder if making an appropraite branch and merging it with --squash is easier to automate19:47
canton7 if you know in advance which commits will be squashed, commit --squash and and rebase --autosquash, or whatever the flag is19:47
angelsl left19:48
joaoh82 joined19:49
panpainter joined19:52
panpainter left19:53
BSaboia joined19:53
joaoh82 left19:53
LucasKA joined19:53
panpainter joined19:53
mar77i joined19:53
panpainter left19:54
dmsuperm1n joined19:54
dmsuperm1n If I have a branch (dev) that somebody else has branched locally, and I rename dev (locally and on github) will there be problems for them when they go to pull?19:55
mar77i_ left19:55
dmsuperm1ndmsuperman19:56
dmsuperman left19:56
dmsuperman joined19:56
kodab left19:56
Guest31032mf_19:57
konadr joined19:58
canton7 dmsuperman, if they haven't pushed it anywhere, typically tracking into won't be set up, so 'git pull' (with no other arguments) wouldn't work anyway19:59
if they checkout dev and 'git pull', that'll fail though ;)19:59
bwreilly joined19:59
dmsuperman Right, presumably they wouldn't notice a difference until they go to merge in which case they'll just have to merge to/from the new branch19:59
ok cool19:59
matteo left20:00
canton7 also, note that remote-tracking branches for branches which have since been deleted remotely aren't pruned until you 'git fetch --prune'20:00
dmsuperman the other thing is will a git pull for a branch deleted on remote (where the default behavior is pull all) -- ah yep20:00
mvelioglu_ joined20:00
canton7 (so origin/dev will hang around for ages in their repo)20:00
shawndumas joined20:00
FortySix_ joined20:00
hs366 left20:01
mvelioglu left20:01
angelsl joined20:02
FortySix2 left20:02
dmsuperman thanks much canton7 for clearing that up20:02
DrShoggoth left20:02
pbrobst left20:02
Targen joined20:03
FortySix2 joined20:03
konadr left20:03
canton7 no problem20:03
dxdt left20:03
canton7 not sure what you mean by 'where the default behaviour is pull all', though...20:04
singingwolfboy left20:04
FortySix2 left20:04
DrShoggoth joined20:04
veebers left20:06
FortySix_ left20:06
petey__ left20:07
thierryp left20:07
sohail joined20:07
dglambert joined20:07
petey_ joined20:07
sohail hey guys, anyone know if there is a way to merge certain directories only?20:08
Houte joined20:08
Nugget make a diff of just the files/directories you want to merge and then apply that with patch20:08
sohail Nugget, when I merge fully later on, will that give conflicts?20:09
joki left20:09
Houte How do I remove a file from the coming commits. I physically removed gameHandler.java. Now it will still be in the repo unless I remove it somehow from git. how do I do that?20:09
Nugget no, because the files will be the same20:09
sohail right20:09
rjc joined20:09
Nugget the downside is that you'll lose any interesting commit logs and development history, since it'll just be one big commit20:09
sohail thanks Nugget20:09
m0ltar joined20:10
petey__ joined20:11
angelsl left20:11
Advocation joined20:11
petey_ left20:11
el3ctron_ joined20:12
syl joined20:13
hwrdprkns left20:13
tirengarfio left20:13
syl !list20:13
gitinfo [!mailing_list] The mailing list can be reached via [email@hidden.address] You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist20:13
syl left20:13
thevishy left20:14
joki joined20:14
cmpera left20:15
thierryp joined20:16
m0ltar left20:18
ColoradoPods left20:18
koda joined20:19
nilg left20:19
eaj joined20:20
tirengarfio joined20:21
m0ltar joined20:21
veebers joined20:22
thevishy joined20:23
LuxuryMode left20:23
thierryp left20:23
msafi joined20:23
kukks left20:25
m0ltar left20:25
codehotter how do I check whether I'm ahead of master or not in a bash script?20:26
dxdt joined20:26
codehotter I mean of origin/master20:26
mklappstuhl joined20:27
Houte How do I remove a file from the coming commits. I physically removed gameHandler.java. Now it will still be in the repo unless I remove it somehow from git. how do I do that?20:27
cjh` codehotter: maybe `git log HEAD..origin/master`, probably a nicer way though.20:27
unsignedint git rm ?20:27
ojacobson codehotter: you probably mean "ahead of upstream"20:27
in which case 'git rev-list --count @{upstream}..HEAD' or similar, depending on use case20:28
reversing the range gets you the 'behind' count20:28
(note that both counts can be non-zero at the same time)20:28
cjh` heh awesome20:28
diegoviola joined20:29
codehotter ojacobson: what is @{upstream}?20:30
ojacobson codehotter: man gitrevisions20:30
gitinfo codehotter: the gitrevisions manpage is available at http://jk.gs/gitrevisions.html20:30
eaj left20:31
davividal hi. On SVN I can checkout my entire repository with branches. If I want to work on branch foo, I can cd into it and work in parallel with branch bar. Is there a way to do something similar on git? I mean: one clone and work on different directories, without git checkout foo|bar ?20:31
ojacobson "mostly" No, git only maintains one work tree.20:32
PerlJam davividal: you can always have multiple clones20:32
ojacobson It's possible (albiet awkward) to set up secondary work trees; there's a contrib script that does this20:32
but a second clone is less work and easier to reason about20:32
danicampa901 joined20:32
cjh` there is git-new-workdir, haven't used it myself though.20:32
davividal "ideally" would be 'one clone to rule them all', but it won't be a real issue to clone N times.20:33
gfontenot left20:33
codehotter refers to the branch that the branch specified by branchname is set to build on top of <-- what does this mean? Is there another man page I can read that would give me an understanding of what "set to build on top of" means?20:33
hobodave joined20:34
ojacobson codehotter: it's the branch listed in 'git branch -vv' as the upstream branch, in [square brackets]20:34
carbon joined20:34
Mars joined20:34
gfontenot joined20:35
ojacobson man git-branch, man git-pull both cover what "upstream" means and how it's specified20:35
gitinfo the git-branch manpage is available at http://jk.gs/git-branch.html20:35
the git-pull manpage is available at http://jk.gs/git-pull.html20:35
ojacobson tl,dr branches aren't necessarily based on origin/master20:35
MarsGuest8082320:35
jason237 joined20:35
hobodave_ left20:35
hwrdprkns joined20:35
carbon left20:36
codehotter ojacobson: I'm trying to detect from the batch script, whether the developer has made any changes since he last ran "git push"20:36
ojacobson oof20:36
there are some unfounded assumptions there, too :)20:36
codehotter uh oh20:36
ojacobson by default (there's a config setting for this) unqualified 'git push' pushes more than just the current branch20:36
dort joined20:36
jackneill left20:36
codehotter well yes, I mean in the current branch20:36
ni__ left20:36
codehotter I mean if the current branch has changed since it was last pushed20:36
ojacobson and the current branch might not be a local copy of the upstream branch -- consider eg. feature branches, which should always have an upstream but should always be ahead of upstream if any work has been done20:36
Luke-Jr joined20:37
ojacobson the upstream of a feature branch should be the branch the feature will land on, not the server's copy of the feature branch, generally20:37
pbrobst joined20:37
ojacobson (it's really a matter of workflow and policy, not technology, but still)20:37
codehotter so lets say a developer has a feature branch, which he pushes to his own repo, and where upstream can then pull from20:37
niels__ joined20:37
dfanjul_ joined20:37
ojacobson yes20:37
the feature branch's @{upstream} is probably origin/master or similar20:37
codehotter I dont care whether upstream has changed, but I want to know if the developer has made any local changes since he last pushed to his own repo20:37
davividal git internals: what happens when I git checkout branch-foo from master? It deletes my files from master and 'copies' from branch-foo?20:37
vince_prignano left20:37
ojacobson _right now_ you can't easily know both20:37
dfanjul left20:38
ojacobson I've heard rumblings that publish information might be added to git-config one day so that you *can* tell whether a local branch is behind its published copy, separate from its upstream20:38
niels__ say a have a develop branch based on master, develop is 100+ commits ahead of master, i don't want anything to do with master, i basically want to have a new master based on develop. how do i do that?20:38
ojacobson but nothing concrete20:38
Landshark753 joined20:38
RaceCondition joined20:38
codehotter ojacobson: so, in short, you can't easily tell whether git push has been run or not?20:39
PerlJam davividal: it makes sure that your working dir contains everything from branch-foo (it may also contain other things however)20:39
ojacobson codehotter: not without additional workflow-specific information, no20:39
PerlJam davividal: so ... kinda what you said except for the "deletes my files' part20:39
frogonwheels left20:40
ojacobson for example, I push branches to a remote named 'owen' but the upstream of most of my branches is 'origin/master'20:40
codehotter ojacobson: that's the workflow we have too20:40
I want to know if 'owen' is up to date, or if local is ahead of 'owen'20:40
cjsarette left20:40
ojacobson you can tell if a local branch is behind its published copy, but only if you know the name of the remote to look at and know that I use the same names locally and on my remote20:40
git doesn't know anything about that and doesn't enforce any of it :)20:40
pbrobst left20:40
codehotter well I can programmatically detect where git push would push to, if you ran it20:40
davividal PerlJam: yeah, but on branch-foo I deleted some lines from README.txt (compared to master). What would happen? What should I read (apart from source code)?20:40
ojacobson can you?20:40
HeN_ joined20:41
codehotter well if you ran it without arugments at least20:41
ojacobson Because in my case it's usually 'nowhere' because of how I have push.default set.20:41
My local branch name is almost never equal to its upstream branch's name, so push bails out because push.default is set to simple.20:41
pheaver joined20:41
ColKurtz left20:41
Kayra joined20:41
codehotter OK, well I know our developers have git push automatically push to their own repos for all their branches. Does this information help?20:41
ojacobson The _default_ configuration is that it will name-match all local branches to branches on the upstream remote of the current branch, and push _all_ of those.20:41
(Yes, this default is mildly crazy.)20:42
HeN_HeN20:42
FortySix2 joined20:43
codehotter in other words they'd have (I think) origin set to their own repo and an additional remote upstream set to upstream20:43
cjsarette joined20:43
ojacobson anyways, if you're willing to make some workflow-specific assumptions in your script, yes, you can check the thing you want to check20:43
sohail left20:43
ojacobson but expecting us to know all of the edge cases in your workflow is, uh, unrealistic :) git's too flexible for that.20:43
cillybabe left20:43
codehotter ojacobson: I realize that now. Thank you for explaining. You've given me a lot of ideas about different possible workflows too20:43
ojacobson You'd be better helped by a stronger understanding of git's underlying theory, so that you can determine how the states you care about will be represented in git20:43
Guest80823 left20:44
nZac joined20:44
codehotter Does a workflow make sense where 'git push' automatically pushes to your own private repo, but 'git status' shows whether you are ahead/behind of upstream? Is that where you set origin to your own repo and configure your branches as tracking upstream?20:45
pbrobst joined20:45
zhenjunyao left20:45
rpg_ joined20:45
rpg left20:46
pheaver left20:47
ojacobson I've got 'git publish' set up as an alias for exactly that20:47
https://github.com/ojacobson/git-aliases/20:47
croaky left20:48
shawndumas left20:49
adam_ I have a situation at work. Specifically regarding submodules.20:49
I'm an advocate of using them because they allow you to have source control of the library you are using20:49
shawndumas joined20:49
adam_ we're a .net shop for the most part but moving into other areas like node.js and golang20:49
joaoh82 joined20:50
panpainter joined20:50
cilly joined20:50
koda left20:50
ojacobson The right way to do dependency management in node.js is probably npm, not submodules20:51
even for "in house" dependencies20:51
I assume golang has something simil;ar20:51
adam_ some have the idea that replacing the use of submodules for reusing our own libraries with a package manager is a good idea20:51
DrShoggoth left20:51
adam_ that's fine for interpreted languages20:51
bartzy left20:51
adam_ C# is compiled and as such, if you use a package manager, you can't step into the code20:51
ojacobson package/dependency managers know about things like transitive dependencies; git does not20:51
abbycabs joined20:51
abbycabs left20:52
adam_ my stance is that it introduces friction as you have to go and explicitly get the library source and set it up to debug20:52
abbycabs joined20:52
adam_ please explain how transitive dependencies are not taken care of by submodules20:52
Turicas left20:52
konadr joined20:53
ojacobson If you include a third-party library using a submodule, you have to convince them to follow your submodule convention for their own dependencies, or manage their dependencies yourself20:53
jason237 left20:53
adam_ further, we'll have to have a separate package manager per technology: nuget for C#, npm for node, and something else for golang - I think it has it's own20:53
ojacobson (most libraries for most languages do _not_ use submodules, so this is an uphill battle)20:54
Yes, that's fairly normal now, sadly20:54
I'd rather that the various lang folks had spent their effort making distro/native packaging work better for their cases, but we have what we have20:54
joaoh82 left20:54
dglambertDGLAMBERT20:54
ojacobson pip, npm, maven, cabal, gems, &c &c &c20:54
DGLAMBERTdglambert20:54
adam_ submodules have been working well for us since 201020:54
croaky joined20:54
konadr1 joined20:55
adam_ so I'm not about to introduce something that will not allow the developers to step into code of the library they are using (in-house lib)20:55
konadr left20:55
Nizumzen joined20:55
ojacobson That seems like such an obvious flaw that I find it hard to believe that you're right (except possibly by default)20:56
m8 joined20:56
adam_ ?20:56
ojacobson Stepping into compiled code from packages is a solved problem for most compiled languages, even C20:56
adam_ it requires us to have 2 builds now.. one with symbols and one without.20:56
ojacobson Does nuget perhaps keep sources/debug symbols separate from the compiled binaries? Maven does this (-sources.jar)20:56
adam_ stuff that gets pushed to production has to have a different compilation20:57
codehotter next question: can I check if there are changes that have not been commited yet?20:57
adam_ this has not been without sideeffects in .net20:57
ojacobson Why? Is there a problem with having debug symbols in production? :)20:57
adam_ yes. PCI complian shop20:57
ojacobson PCI does not make any statements about debug symbols.20:57
codehotter PCI mandates that you do not have debug symbols in production? Or pci mandates compile options that you do not want to use in test?20:57
ojacobson (If it did, it would also be nearly impossible to use interpreted languages in PCI-compliant contexts, since source is strictly more informative than debug symbols.)20:58
harobed joined20:58
ojacobson codehotter: of course you can.20:58
adam_ I think PCI implies make it as hard as possible for a hacker that got in if it's a card holder env20:58
codehotter ojacobson: how?20:58
ojacobson adam_: anyways, as you can tell, I think that using submodules to manage dependencies is not worth the theoretical simplicity20:58
petey__ left20:58
guardian why are there git commands implemented as shell scripts that don't use git-rev-parse? because they existed before git-rev-parse and its parseopt mode?20:58
codehotter ojacobson: I mean from a bash script. Obviously a user can get with git status20:58
daivyk left20:59
ojacobson codehotter: so why can't your script? Hint: man git-status20:59
gitinfo codehotter: the git-status manpage is available at http://jk.gs/git-status.html20:59
pheaver joined20:59
adam_ the other problem with not using submodules is that we now have to host nuget for just dependency management20:59
or open up to microsoft20:59
konadr1 left20:59
ojacobson From my own PCI experiences, I would have thought you'd have to do that either way, in the process of satisfying requirements around reproducibility and handoff.21:00
quetz left21:00
alpha123 joined21:01
a-l-e left21:01
adam_ anyway, I still don't get your transitive argument. I'm strictly talking about using in-house libraries which we want to change quite frequently.21:01
Hydrant left21:01
adam_ going through a package management publishing workflow only adds friction21:02
msafi left21:02
beyondcreed left21:02
simesy joined21:02
adam_ so why would you not use submodules?21:02
codehotter ojacobson: I can use LOCALCHANGES=$(git status -s) and then test if this is empty or not. Thanks.21:02
ColKurtz joined21:02
ColKurtz left21:02
petey__ joined21:02
codehotter I do try to read the manpages before I ask the question, but when I get pointed at the same manpage again, I always see things I didn't the first time21:02
ojacobson adam_: One, transitive dependencies; I don't feel like working around each developer's personal preferences on how to express their own deps as submodules (or to not do so)21:02
Two, I don't feel like rebuilding/repackaging everyone else's code21:03
ColKurtz joined21:03
ColKurtz left21:03
ojacobson especially in the presence of special-snowflake build processes (which are pretty widespread even in Python)21:03
Advocation left21:03
adam_ those are 0 friction21:03
ojacobson Lucky you.21:03
Mangellan2497 left21:03
ojacobson May that remain true for you.21:03
adam_ since 2010 it has21:03
ColKurtz joined21:03
ojacobson It's never been true for me or anyone else I know; anecdote is not data, but that's an experience I'll bank on :)21:04
ncthom91 left21:04
ojacobson 2010 isn't very long.21:04
adam_ it will be 4 years strong this summer21:04
ojacobson Three, the submodule interface is awful21:04
adam_ we have 80 repos21:04
cilly left21:04
LBV_User left21:04
adam_ so part of why I'm asking is that I do not want to break what isn't broken21:04
Advocation joined21:05
ojacobson Rather than defending the existing system, why not make the people proposing package management justify the change?21:05
You're more likely to be convinced, that way.21:05
adam_ I agree that the submodule interface is pretty bad21:05
panpainter left21:05
ojacobson (Or not, as the case may be.)21:05
tpollard joined21:05
panpainter joined21:05
ojacobson If you're moving into node, point one will bite you pretty hard though21:05
Node packages *pervasively* require other node packages21:06
cilly joined21:06
mar77i_ joined21:06
adam_ that's fine for 3rd party21:06
ojacobson 'npm install karma' installs ... /counts/21:06
adam_ I don't things its fine for inhouse21:06
it's not a big shop21:06
cbreak why do people hate submodules so much?21:06
mar77i left21:07
adam_ about 25 devs colocated so if there's an issue, it gets resolved very quickly21:07
cbreak they are excellent for what they are intended21:07
adam_ usually it's lack of knowledge21:07
and lack of using the command line in the first place21:07
ls___ joined21:07
cbreak I'd guess it's the cause of most of the problems in this channel: Tool Abuse21:07
adam_ ... especially in .net circles21:07
mar77i_mar77i21:07
adam_ yes, we have 1 abuse of submodules21:07
cbreak like abusing submodules to split a repository you actually develop on.21:08
or that guy yesterday who wanted a recursive submodule...21:08
adam_ we use common code that should be running as a service, as a library21:08
milki its submodules all the way down?21:08
adam_ we have 3 levels at some point21:08
this is where they are abused21:08
normally it's only 2 levels21:09
koda joined21:09
adam_ this is fine as we usually do work on both levels, so submodules do add the sanity check just to see where you are before you commit to using a particular version of the shared code21:10
simesy left21:10
adam_ adding a package manager like nuget makes it hard to work on both at the same time21:10
mdhopkin_ joined21:10
Nizumzen left21:10
cilly left21:11
milki package managers should not be used during development o.O21:11
adam_ ?21:11
cilly joined21:11
adam_ well, this is what they are advocating21:11
milki thats...interesting21:11
adam_ for in-house, currently worked on libraries21:11
milki how do they intend to put package managers into the developer workflow?21:12
ls___ Perhaps somebody knows how to merge, a single file from sha1 sum?21:12
cbreak ls___: nope.21:12
ls___: you can get the file though and merge it yourself21:12
infynyxx left21:12
cbreak ls___: or you can use git checkout -p commithash -- file21:13
ojacobson ls___: git is built around the premise that you merge whole lines of history, not single files or single changes21:13
cbreak make sure that you have no uncommitted changes in the working dir21:13
also, this will not be a merge21:13
it'll just give you the changes21:13
ojacobson but given a line of history that includes changes to a file, git can help you build a second line of changes that _only_ includes changes to that file21:13
da5id left21:14
cbreak that's kind of cheating though.21:14
Guest12060 left21:14
Guest12060 joined21:14
Guest12060gnarf21:14
axl_ left21:14
milki cheating is fun21:14
Nevik milki: simple: "Guys, this is Sam. Sam will be your package manager. Whenever you build, you tell Sam. He'll be here and take care of you."21:14
ls___ `Right, I'm trying to understand a simple workflow21:14
The git scm book has been helpfull at the same time.21:14
milki sam: f this. *flips tables* _all_ the tables21:15
Nevik lulz21:15
that escalated quickly21:15
cbreak yeah, who needs package managers?21:15
I just use cmake to find the deps, and list the dependencies in the readme file.21:15
milki how does cmake do that?21:15
cbreak find scripts.21:15
Nevik cmagic!21:15
cbreak it has a pkgconfig backend too21:16
pantsman left21:16
milki shiny21:16
cbreak on windows it doesn't work too well21:16
you have to set env vars usually21:16
but that's because that platform doesn't even have a standard lib location...21:16
on OS X it finds libraries in /usr/lib, /usr/local/lib and so on quite automatically21:17
adam_ So the thing that is bothering me about introducing a package manager and dropping submodules as the way to manage the dependency is that extra friction when you want to work on both pieces of code at the same time21:17
ls___ ... automagically21:17
jooools left21:17
adam_ even if you're not, it's faster to find a bug and send the potential fix/do the fix and pass off to the team that owns that library21:17
mizu_no_oto left21:18
milki adam_: i dont expect package managers to be used during the development process, only testing/staging/deployment21:18
adam_ I see a lot of steps being added to do the same thing with a package manager playing a safety net in the middle, effectively turning the library that your dev team wrote into a black box21:18
milki: even then, why bother?21:19
hoschi left21:19
milki because its an existing infrastructure that works for the target OS21:20
adam_ just compile it with the top level code21:20
milki if you build one yourself, then it doesnt matter as long as you automate the process21:20
cwillu_at_work left21:20
pantsman joined21:20
pantsman left21:20
pantsman joined21:20
adam_ yes, the build server does that21:20
cwillu_at_work joined21:20
milki do you use the build server in testing/staging/deployment now?21:20
adam_ so when you ship the top level code, you ship the libs it depended upon21:20
we use teamcity to build the artefacts21:21
milki do you use the build server in testing/staging/deployment now?21:21
adam_ the artefacts get a final runthrough our test suite and then those same artefacts get deployed if the tests pass21:21
ojacobson Hmm, what does '120000' mean in the flags column of git ls-tree?21:22
adam_ the build server is in the dev environment21:22
milki then, it doesnt seem like theres a general need to change your system21:22
adam_ the artefacts get shipped to production and/or staging21:22
milki you should have them argue their viewpoints21:22
adam_ that's what I thought21:22
calcifea left21:22
adam_ they are.. it's about control21:22
b3tamax left21:22
adam_ they don't want some other dev messing with their library21:23
milki then they have other problems to deal with21:23
adam_ they can reject others' chanes21:23
exactly21:23
those are way worse problems21:23
ojacobson The ability to reject changes is fundamental to having working code review21:23
milki control is usually a bad thing21:23
ojacobson it's also fundamental to broken social ecosystems21:23
take your pick :)21:23
grawity ojacobson: '12' is a symlink – see stat(2)21:23
milki hahahaha21:23
ojacobson grawity: aha, thanks21:23
adam_ and putting a package manager in front to make it difficult for someone to change code is such a nasty bandaid21:23
milki ya, thats not a good reason -.-21:24
adam_ It's one guy's reason and he's senior21:24
milki but he's senior...in .net21:24
adam_ the problem is that juniors are jumping on board because that's what all the cool kids in .net shops use21:24
well, yes and no21:25
IrishPrime left21:25
adam_ he's senior in a lot of stuff and is generally good for such things as Domain Driven Design (see Eric Evans), CQRS and Event Sourcing21:25
however, he's very quick to throw in arbitrary stuff without regard to side effects to get stuff done21:26
phipes joined21:26
juliohm joined21:26
jason237 joined21:26
khmarbaise joined21:27
adam_ so I'm trying to do a LOT of educating in a short amount of time21:27
bartzy joined21:27
salamanderrake left21:28
eaj joined21:28
adam_ so here are some submodule issues I'd like to resolve:21:29
mvensky joined21:29
adam_ 1. giant 3rd party dll storage repo. Slows down the build server.21:29
ls___ Is this the correct command to get the log of a particular branch 'git log <branchname>?21:30
milki ls___: !tias21:30
gitinfo ls___: Try it and see™. You learn much more by experimentation than by asking without having even tried. If in doubt, make backups before you experiment (see !backup). http://gitolite.com/tias.html may help with git-specific TIAS.21:30
adam_ ls___: yes21:30
harobed left21:30
ncthom91 joined21:30
irqq joined21:30
alpha123 left21:31
milki build server should be smarter in grabbing the correct dlls21:32
storing them in a git repo sounds like a bad idea (tm)21:32
Advocation left21:33
sq-one left21:33
salamanderrake joined21:34
thierryp joined21:34
dinoangelov joined21:34
marcjs joined21:34
ls___ left21:35
marcjs is it possible to squash foreign commits?21:35
dreiss left21:35
_ikke_ foreign commits?21:35
milki you can manipulate commits that you have reference to locally21:35
marcjs I hate it to bother our contributors in their PullRequest at github with squashing commits.21:35
jerius joined21:35
gdogrady joined21:36
eaj left21:36
Luke-Jr left21:36
milki marcjs: https://help.github.com/articles/merging-a-pull-request#merging-locally21:36
do it locally21:36
Luke-Jr joined21:37
ncthom91 left21:37
Mars___ joined21:38
marcjs milki: yeah I know how to merge manually, but when I squash other's commits won't I appear then as the author?21:38
Bass10 left21:38
canton7 marcjs, git differentiates between author and committer21:38
rpg_ left21:38
maestrojed I wanted to rename a branch. Locallly I did "git checkout oldName" and "git branch -m newName". I pushed newName to origin which went fine. But origin/oldName still exists. What did I do wrong?21:39
thierryp left21:39
maestrojed or how do I fix it21:39
canton7 marcjs, so if you rebase some files, you become the committer, but the author's unchanged21:39
hwrdprkns left21:39
phile left21:39
marcjs ok good to know, then it should work :) thanks!21:39
rpg joined21:39
canton7 maestrojed, nothing. remove the old branch from the remote with 'git push --delete remotename oldName'21:39
slyfox left21:39
milki maestrojed: you need to explicitly delete the old branch from the remote21:39
maestrojed canton7 milki great thanks!21:40
mklappstuhl left21:40
geekbri_ left21:40
krkinikini21:41
brisbin left21:41
yiati joined21:42
marhaban joined21:42
yiati How unsacred would you all say it is to tinker with the source code of a submodule21:42
zapb_ left21:42
amortimer_mbp left21:43
sgambino left21:43
eijk_ joined21:43
infynyxx_ joined21:43
canton7 yiati, "source code of a submodule"?21:44
i.e. 'cd submodule; do stuf....'?21:44
brisbin joined21:44
FatDarrel_ joined21:45
yiati i.e. Modifying an external project that you are using for your project slightly because it doesn't do quite what you want it to do, thus creating your own local branch that you have to maintain yourself21:45
jonatasnona left21:45
adam_ yiati: I'm very happy with it21:45
you are effectively forking21:46
and it's very powerful21:46
yiati No one has any issues doing this, or views it as hacky?21:46
adam_ nope21:46
ncthom91 joined21:46
cilly left21:46
yiati You have to maintain your own version of an external project21:46
canton7 you can even submit your patches back to the original maintainer :)21:46
adam_ if you generalize the solution, you can make a pull request21:46
yiati And completely lose all benefits of the open source community21:46
canton7 yiati, no? you can still merge in changes from upstream21:47
eijk left21:47
adam_ if it's accepted you can go back to using the core again21:47
yiati canton7: I'm not talking about pull requests, I think those are fantastic21:47
sgambino joined21:47
adam_ yiati: you can still keep up to date and your changes at the same time21:47
just keep merging or rebasing based on your preference21:47
FatDarrel left21:48
FatDarrel_FatDarrel21:48
adam_ don't forget about rerere if your conflicts are starting to bother you21:48
mvelioglu_ left21:48
canton7 but if you're including that project as a submodule, then you'll have to manually update once in a while anyway21:48
rfoust left21:48
yiati that sounds horribly unnecessary. The real solution should be to fix what's wrong on your end, not in the other projects source code21:48
Even if it takes a bit more code21:48
Makes it much easier to maintain21:48
jleclanche left21:48
canton7 sure, but that's not always possible21:49
yiati :(21:49
guardian why are there git commands implemented as shell scripts that don't use git-rev-parse? because they existed before git-rev-parse and its parseopt mode?21:49
canton7 fixing original > working around in your own code > maintaining fork with no intention of submitting back to upstream, IMO21:49
yiati Alright I can at least agree to that ordering21:50
I just feel like maintaining a fork is an ugly last resort that some people willingly jump onto if it's the easier option21:50
joaoh82 joined21:50
da5id joined21:52
milki guardian: you said that already21:52
robbyoconnor left21:53
marthinal left21:53
infynyxx_ left21:53
cilly joined21:53
dreiss joined21:53
ojacobson guardian: why not ask their respective authors?21:53
robbyoconnor joined21:53
therealGent joined21:54
croaky left21:54
milki or check the commit dates21:54
joaoh82 left21:54
croaky joined21:54
ojacobson (This is largely a chatroom full of git users; asking git development history questions here is maybe less productive than you'd hope. Try the mailing list, where most of the active git developers hang out. :)21:55
guardian I'm looking for some guidance whether it should be used by new commands or not :(21:55
milki you should have said so in the first place -.-21:55
guardian actually some commands use OPTIONS_SPEC, and git-sh-setup tests if OPTIONS_SPEC is defined but never mentions it in its manual21:55
simesy joined21:55
guardian and the man for git-rev-parse mentions OPTS_SPEC as an example21:55
hwrdprkns joined21:55
AaronMT left21:56
jcvfen left21:56
pheaver left21:57
mizu_no_oto joined21:57
guardian but not OPTIONS_SPEC so it's a bit rough to connect the dots :)21:57
milki guardian: you probably find the git mailing list more productive since actual git developers are there21:58
guardian yep I'll collect my thoughts and questions and try there21:58
jcvfen joined21:58
milki again, make sure to explain what you are trying to do before diving into your questions first21:58
endiruna joined21:59
Freeman42 left22:01
cannonball left22:02
dort left22:03
gfontenot left22:03
nZac left22:03
emmajane[emmajane]22:04
dougsland left22:04
dougsland joined22:05
bl33 left22:05
Kayra left22:05
infynyxx joined22:06
Kicer86 left22:06
bdavenport left22:06
nZac joined22:06
jnavila left22:08
pheaver joined22:09
hwrdprkns left22:10
xiong Please talk to me about this kind of subtree: https://github.com/apenwarr/git-subtree/blob/master/THIS-REPO-IS-OBSOLETE22:10
Pullphinger joined22:10
cilly left22:10
grawity https://github.com/git/git/tree/master/contrib/subtree22:11
xiong I have installed git from wheezy-backports which gives me git version 1.8.5.2; but $ man git subtree fails. Also, hard to find out about this browsing the web; too many ambiguities in the word 'subtree'.22:11
gitinfo the git manpage is available at http://jk.gs/git.html22:11
alezost left22:11
yhager joined22:12
cagedwisdom joined22:12
cilly joined22:13
xiong I'll need a tad more context there, friend. Can I obtain this feature within any Debian package?22:13
xlinkz0 left22:13
jrnieder xiong: if you want to help with packaging, sure22:13
xiong Oh dear.22:13
grawity xiong: git-subtree is in contrib/, so it's probably somewhere in /usr/share/git/subtree/22:13
jrnieder xiong: see http://bugs.debian.org/70465222:13
xiong The main git manpage is nothing.22:13
jrnieder xiong: gitinfo is a bot22:14
xiong: it just did some pattern matching on your question and produced a not-too-helpful answer22:14
xiong So perhaps git-subtree is rattling around in my system now but I need to locate it, maybe move it, and locate its manpage?22:15
msafi joined22:15
msafi left22:15
msafi joined22:16
cjwilbur_ left22:16
xiong I thought I had a case that demanded submodules and I tried them. Now I am chastened and will use subtrees if I can.22:16
However the only replacement offered for what's described as a nasty feature is... difficult to get.22:17
EugeneKay !subrepos22:18
gitinfo [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.22:18
EugeneKay My favorite solution is none of this - take a tarball of the subrepo and just unzip it into your repo; git add path/to/lib/ && git commit -m 'Add FooLib v1.2.3'22:18
xiong http://git-scm.com/book/en/Git-Tools-Subtree-Merging22:19
eletuchy_ left22:19
zumba_addict hi folks, I have 2 local branches, can I combine then together?22:19
EugeneKay zumba_addict - man git-merge22:19
gitinfo zumba_addict: the git-merge manpage is available at http://jk.gs/git-merge.html22:19
zumba_addict cool22:19
eletuchy joined22:19
zumba_addict looks like it's the same approach i did earlier22:19
cilly left22:19
zumba_addict i git pulled master, then i git checkout my-local, then I git merge master22:20
pbrobst left22:20
dprophit left22:20
xiong EugeneKay, I seem to have difficulty expressing my concerns. I'm dealing with a number of subprojects; most of them are, indeed, projects in their own right. I need to be able to deal with them as such.22:20
That is, I need to be able to navigate into the subproject, checkout or update, pull or push.22:20
vince_prignano joined22:20
xiong Push to a fork and submit a pull req upstream.22:21
cilly joined22:21
xiong These foreign projects are not going to alter anything to accommodate a wrapping superproject.22:21
jleclanche joined22:21
osmosis left22:22
matanya left22:22
corni left22:23
eletuchy left22:24
xiong I can't use any approach that discards the link to upstream. Is that not a sane statement?22:24
mizu_no_oto left22:24
krononaut left22:24
d0liver__ left22:24
hellopat left22:24
Kayra joined22:24
kermit left22:25
simesy left22:25
jerius left22:26
zumba_addict i need to switch temporarily to a local branch but there are conflicts on my other branch. How do I swithc?22:26
infynyxx left22:26
xiong git-subtree looks, on first reading, to be a sane solution. Why is it hidden away?22:26
zumba_addict i need to see a file22:26
breakingthings left22:26
grawity zumba_addict: a) `git stash`, then switch22:26
RaceCondition left22:26
grawity zumba_addict: or b) `git show otherbranch:path/to/file`22:26
zumba_addict then later git stash pop22:26
i like the latter22:27
thanks22:27
cilly left22:27
mvelioglu_ joined22:27
kermit joined22:28
psyl0n left22:28
jason237 left22:30
xiong I have added a 'me too' to that thread.22:30
d0liver_ joined22:31
Guest34984 joined22:31
Guest34984 hi22:31
gitinfo Guest34984: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.22:31
Guest34984 getting some error at github. error: src refspec master does not match any. error: failed to push some refs to 'https://github.com/username/otb.git'22:31
RaceCondition joined22:32
Guest34984 git push -u origin master22:32
d0liver_ left22:32
leeN left22:32
d0liver joined22:33
grawity Do you have a branch named 'master'?22:33
cilly joined22:33
bubbleheadinc left22:33
eaj joined22:33
Guest34984 I just followed the github post repository creation22:34
grawity, i think the github is implicitly named remot22:34
zumba_addict git stash didn't work when there is a git merge failure22:34
Guest34984 grawity, i think the github is implicitly named remote master22:34
calcifea joined22:35
ojacobson Guest34984: what's currently true on github has almost no bearing on the message you recieved.22:37
shawndumas left22:37
ojacobson Only what's currently true in your local repo can affect that message.22:37
Do you have a *local* master branch? What does 'git branch' say about it?22:38
DolpheenDream_ left22:38
zumba_addict how do I find out if there are still conflicts?22:38
git status only shows the changes22:39
found it by using find22:39
niatish left22:39
Landshark753 left22:39
shinnya joined22:39
cilly left22:39
axl_ joined22:40
ojacobson left22:40
hwrdprkns joined22:40
leandrosansilva_ joined22:40
Guest34984 grawity, http://pastie.org/861206822:41
cilly joined22:41
WaReZ left22:41
Klagid_ joined22:41
leandrosansilva left22:41
eletuchy joined22:41
shawndumas joined22:41
Klagid_ left22:42
eaj left22:42
master_j left22:43
tang^ Guest34984: github says the URL you're pushing to does not exist. did you create the repo on github first?22:43
GT4066 joined22:43
Guest34984 yes..22:43
tang^ oh wait, you never actually committed anything22:44
you need to read the message22:44
bamj0rb Guest34984: Your branch does not exist until you commit something.22:45
Guest34984 yes22:45
how to commit22:45
tang^ nothing added to commit but untracked files present (use "git add" to track)22:45
bamj0rb Use `git add' to add files to the index. Use `git commit' to commit them.22:45
Guest34984 well i did git commit -m "first commit"22:45
tang^ and it said 'nothing added to commit but untracked files present (use "git add" to track)'22:45
frogonwheels joined22:45
FerreusR left22:47
livingstn left22:47
Guest34984 working now. thans . i did add *22:47
Silmarilion joined22:47
yiati left22:48
mhi^ joined22:49
bartzy left22:49
mdhopkin_ left22:49
bamj0rb You probably wanted . instead since * would (in sane shells) glob to non-hidden files, which may not be everything you want. You can also just say -A.22:49
cilly left22:50
infynyxx joined22:50
cilly joined22:50
gdogrady left22:51
Landshark753 joined22:51
mdhopkin_ joined22:51
joaoh82 joined22:51
sgambino left22:52
abbycabs left22:52
grawity left22:52
grawity joined22:54
tang^ left22:54
gdogrady joined22:55
cilly left22:55
bitslayer joined22:56
joaoh82 left22:56
konadr joined22:56
pheaver left22:57
amcsi joined22:57
konadr1 joined22:58
konadr left22:58
diegoviola left22:58
zhost left22:59
m-o-s joined23:00
thiago joined23:00
sgambino joined23:01
konadr1 left23:02
aard__ left23:02
pinage404 left23:03
tieTYT left23:03
Diemuzi joined23:04
gitinfo set mode: +v23:04
dmsuperman left23:04
Dougie187 left23:04
[emmajane]emmajane23:04
tieTYT joined23:04
mvelioglu_ left23:05
m-o-s left23:05
hjp_beta__ left23:05
croaky left23:05
dreiss left23:05
harleypig joined23:06
rendar left23:06
RobOakes joined23:06
petey__ left23:07
RobOakes Is there a way to force git to ask for my username/password when updating a bunch of submodules?23:07
mfulz left23:07
juliohm left23:07
RobOakes I'm not terribly excited to put in my username/password into .gitmodules, as the file is under version control.23:07
danecando joined23:07
petey_ joined23:07
dxdt left23:07
RobOakes And I can't figure out how to have git reliably prompt for the username/password (I'm using http, not ssh).23:08
pantsman left23:08
grawity man gitcredentials23:08
gitinfo the gitcredentials manpage is available at http://jk.gs/gitcredentials.html23:08
pantsman joined23:09
pantsman left23:09
pantsman joined23:09
eaj joined23:09
croaky joined23:10
webguynow left23:11
webguynow joined23:12
dxdt joined23:12
khmarbaise left23:12
petey_ left23:12
subsonic joined23:14
longbeach left23:14
irqq left23:14
niels__ say a have a develop branch based on master, develop is 100+ commits ahead of master, i don't want anything to do with master, i basically want to have a new master based on develop. how do i do that?23:15
arand__ niels__: do they diverge?23:16
Kayra left23:16
dort joined23:16
subsonic When I branch from staging -> to branchtest, commit, then merge staging into branchtest (to look for updates), switch to staging, then merge branchtest into staging with --no-commit it doesn't follow the no-commit and automatically merges. Anybody know why?23:16
dreiss joined23:17
osse niels__: suppose you could magically move the 'master' "label" from where it is to where 'develop' is now. Would that be awesomesauce?23:17
pheaver joined23:17
Klagid_ joined23:18
eaj left23:18
mdhopkins left23:19
SJrXSJr23:19
panpainter left23:19
joshc subsonic: is it performing a fast-forward "merge"? in that case no merge commit is created23:19
FortySix2 left23:20
thevishy left23:21
osse niels__: I'mn going soon so in case the answer to my question is "yes" here's the procedure for that: git checkout master; git reset --hard develop.23:22
daivyk joined23:22
diegoviola joined23:22
torbjorn_ joined23:24
niels__ arand__: yes, i need to get completely rid of master. i just want a copy of develop as master.23:24
pbrobst joined23:24
torbjorn_ I easily do git tag help instead of git help tag, same with branch. how about making git prompt "are you sure" when ever objects named "help" are created where it might not be right?23:25
mdhopkin_ left23:25
Klagid_ left23:25
d0liver niels__: "master" is just a pointer. you just need to move it to the tip of develop like osse said.23:26
Guest34984 left23:26
vince_prignano left23:26
nZac left23:26
osse \o/23:27
niels__ d0liver: so if i do as he says, the current master will no longer be, and it'll be a copy of develop so to speak23:27
d0liver niels__: sort of23:27
your current master won't have anything pointing at it23:27
and it will eventually be cleaned up if left alone long enough23:27
akersof left23:28
d0liver but really all you're doing is moving the pointer that is "master" to the tip of develop23:28
dxdt left23:28
bamj0rb left23:28
Landshark753 left23:29
parcs hello23:29
osse niels__: I like to think of it as if you drew the history on a whiteboard in the office with two post-it notes on it. You then move the 'master' post-it note to where the 'develop' post-it is. And after a while the cleaning lady wipes a way the part that used to be 'master' because it's just sticking out to the side :p23:29
niels__ ahhh23:30
that makes sense23:30
kill-9_ left23:30
dxdt joined23:31
osse that way of thinking about it was taken quote literally from one those videos23:31
!videos23:31
gitinfo [!talks] Some good video talks about Git: [yt] http://goo.gl/z72s (Linus Torvalds: History&Concepts); [yt] http://goo.gl/R9H2q (Scott Chacon: Git basics, live examples); http://vimeo.com/35778382 (Randal Schwartz: Git basics, descriptional); http://vimeo.com/46010208 (Jesica Kerr: Git basics, descriptional)23:31
osse The last one23:31
s/quote/quite/23:32
xiong This does not seem to apply to my situation; I have installed git from Debian wheezy-backports: https://github.com/git/git/blob/master/contrib/subtree/INSTALL23:33
Mangellan2497 joined23:33
xiong I'd like to use this feature, really. Advice?23:33
dfanjul_ left23:33
Landshark753 joined23:33
_Qman left23:34
kodab joined23:34
xiong Do you think this really needs to amount to more than sticking git-subtree.sh somewhere in my $PATH? Ideas for a better approach?23:34
simesy joined23:34
d0liver What's the problem with the existing approach?23:36
Error404NotFound joined23:36
panpainter joined23:36
Dougie187 joined23:37
koda left23:37
osse niels__: if develop is literally ahead of master (ie. master is somewhere in the history of develop) then you can do a fast-forward merge. The result would be the same. In this case though it seems the idea is to nuke master from orbit so reset --hard probably feels more right :P23:37
Destos left23:38
msafi left23:38
niels__ osse: Develop-> This branch is 127 commits ahead and 3 commits behind master23:40
Splenetic left23:40
VictorCL joined23:40
petey_ joined23:41
d0liver xiong: It kind of looks like you're supposed to be rebuilding git for that tool23:41
osse niels__: ahh ok. so then it will be sticking out to the side and the cleaning lady will fix it23:41
SocalNick left23:41
danecando left23:41
niels__ osse: haha, love that23:41
osse: which.. is.. git checkout master; git reset --hard develop, right?23:41
osse niels__: oui, monsieur23:42
nnyk_ joined23:42
imachuchu niels__: right, but what about those "3 commits behind master"? It sounds like master has 3 commits that aren't on develop23:43
grawity yes, that's what it menas23:43
means*23:43
halu left23:43
osse imachuchu: those are bastard children than niels__ wants nothing to do with23:43
d0liver yeah, he said he was okay with that23:43
he said he doesn't care about master23:43
niels__ exactly, the master should vanish!23:44
d0liver I would still probably git checkout -b master-old first23:44
if it wre me23:44
*were23:44
mitz_ left23:44
lavaman left23:44
imachuchu ok, Just wanted to make sure that he knew that there were 3 commits that were going to be lost23:44
osse d0liver: but then you have to checkout master again. git branch master-old :)23:44
SocalNick joined23:45
Noldorin left23:45
d0liver it's still the same # of commands I think23:45
if you start on develop23:45
actually23:46
jdav_gon` joined23:46
akersof joined23:46
d0liver no, that's weird23:46
you would want to start on master23:46
meh23:46
EugeneKay xiong - listening to your description, why not just use independent repos? Why the need for a "superproject" at all?23:46
d0liver you win23:46
xiong d0liver, I think it's called upon but not actually required -- I don't know.23:47
EugeneKay (went to lunch)23:47
d0liver xiong: It's probably required because you're going to type things like git subtree .. blah23:47
which means you need to extend git23:47
unless you want to use the hyphen everytime23:47
xiong EugeneKay, because although there will be some work done on individual projects, the main demand is to integrate them into a working whole.23:47
Fiouz_ left23:47
Kayra joined23:47
mlimaloureiro joined23:47
halu joined23:48
leandrosansilva_ left23:48
d0liver I'm inclined to agree with EugeneKay though because subtrees are sort of non-standard and they support bad workflow I think23:48
jdav_gone left23:48
leandrosansilva joined23:49
EugeneKay Is this for the purpose of releasing a "working" codebase?23:49
d0liver It sort of encourages you to have a very fuzzy notion of what belongs in this project and what belongs in others23:49
Fiouz joined23:49
xiong I may seem to be contradicting myself but that's just because the resolution will be different in different cases. A, drop in the subproject and it just works. B, fight it out and toss the subproject as conflicting with everything else. C, modify one sub to play nice with the rest. D, add a third sub that makes one of the two fighters get along with the rest. And so on.23:49
Tell me how I can otherwise deal with all these weird cases.23:50
aberrant left23:50
xiong The project looks different when viewed from different angles. I can't help that.23:50
Pullphinger left23:50
rpg left23:51
xiong I must have a clean, trackable way to add projects into the big mess, remove them, and checkout different versions of any given one.23:51
akersof left23:51
d0liver xiong: Use a dependency manager like composer23:51
dxdt left23:51
mlimaloureiro left23:52
d0liver this isn't really a git thing unless you're working on those projects simultaneously23:52
joaoh82 joined23:52
d0liver and if you're working on those projects to the extent that you can't make your modifications in your local project23:52
Kayra left23:52
d0liver then make your own fork23:52
and use your dependency manager to pull from your fork23:52
xiong In what way am I *not* working on them simultaneously?23:52
VictorCL left23:53
d0liver You shouldn't be changing the codebase of all the projects you're using23:53
think about what a nightmare that's going to be when they update23:53
xiong No. I should only change a sub when necessary.23:53
joaoh82 left23:53
d0liver but it's always going to be necessary to get updates, bug fixes, etc23:53
and when you do it's going to be terrible23:53
xiong From upstream's viewpoint they should be able to look at a simple pull request from a fork of their project alone.23:53
decotii left23:54
joaoh82 joined23:54
Houte left23:54
d0liver then make a separate repo23:54
and send a pull request from that one23:54
that way your changes to that project are localized23:54
but in general, if you're not planning on contributing your changes23:54
akersof joined23:54
xiong Of course, d0liver. That's what I intend.23:54
d0liver then you shouldn't be making them in the codebase for that project23:54
if you're planning on contributing23:54
xiong The super is an integration device.23:54
d0liver then get a dependency manager23:54
and hook it up to your forks23:55
then update your forks23:55
and use the dep manager to pull from your forks23:55
xiong Okay. So what you are saying is that I don't want git to control the superproject at all??23:55
d0liver Eh. no.23:55
I'm saying have git control the "superproject"23:55
VictorCL joined23:55
d0liver and figure out what your subprojects are23:55
create a repo for each of those23:56
Exancillatus joined23:56
d0liver then use a dependency manager to pull those into your super project23:56
that way your subprojects have a life of their own23:56
and they're not tied to your super project23:56
xiong Okay, I obviously need more study-up. 'composer'?23:56
d0liver it will be much cleaner in the end23:56
therealGent left23:56
d0liver composer is a php thing23:56
but a dependency manager of some type23:56
simesy left23:57
d0liver which basically just pulls in repositories into your repo23:57
xiong Right!23:57
d0liver you specify a version and it creates a lock23:57
mitz_ joined23:57
xiong That's what I've been trying to do with submodules and subtrees.23:57
d0liver and it will manage updates for you etc.23:57
Yeah, but submodules and subtrees are half baked23:57
you need a different tool23:57
joaoh82 left23:58
xiong Well, not to be rude; but this might be better advice to give at the outset, instead of saying 'Don't use submodules, use subtrees.' Not saying you personally said that, either.23:58
VictorCL left23:58
d0liver well, submodules and subtrees can serve a purpose too.. but they're kind of weak tools imo23:59
xiong I'm unaware of the existence of such a... super-git, if you will. Let me do study-up; I have much to learn.23:59
d0liver okay23:59
Noldorin joined23:59
hyperair left23:59

Logs Search ←Prev date Next date→ Channels Documentation